Lightning risks and mitigation solutions for vessels

1. Introduction

Thousands of thunderstorms are in progress at any given time throughout the world, resulting in tens of millions of lightning strikes to Earth each year [1]. Climate change modelling shows that global lightning activity is predicted to increase by about 12% for every degree of rise in global average air temperature [2], so the frequency and severity of thunderstorms is also in-creasing.

Whilst total lightning activity over bodies of water (oceans, lakes and rivers) is typically less than the activity over the same area of land, there are several aspects of thunderstorms over wa-ter that make lightning extremely dangerous. For example:

• (a) Pre-strike electric fields are 4-5 times larger over water than over land [3].
• (b) Lightning strikes are more intense over oceans [4].
• (c) Isolated objects on water, such as ships, become the most likely strike point for lightning due to the lack of compet-ing points that initiate upward leaders – a key element of the lightning attachment process that determines the strike point [5].
• (d) The isolation of being at sea means that equipment in ves-sels is “mission critical” and injuries to occupants is more “catastrophic”.
• (e) Lightning strike density in shipping lanes is higher than similar yet relatively unnavigated areas of the ocean [6].

In point (a), the measure of “intensity” depends on the method of observation, e.g., optical vs radio sensors at ELF, VLF, and VHF. However, lightning flashes over oceans have larger cur-rents, longer durations, and are brighter than flashes over land [4]. Also, Holzworth et al. [7] found “superbolts” with energies that are 3σ above the mean occur predominantly over oceans.

In point (b) and leading into point (c), unlike structures on land, any vessel on water presents as a very likely strike point in a thunderstorm due to its electric field enhancement with little “competition” from adjacent points. When lightning enters a ves-sel, it tries to find a low-impedance path to the water. Hence, it will damage whatever comes between it and the water.

Damage commonly seen in lightning strikes to vessels include:

• • Electrical failures – electrical components such as the battery, refrigeration controls, air conditioning, instru-ments, sensors, and controls.
• • Mast damage – particularly sailboats with non-metallic masts.
• • Hull problems – particularly those made of fibreglass, but a hole in any hull is possible.
• • Catastrophic events – lightning is well-known to start fires on unprotected vessels.

Of course, personnel safety is also compromised because there is a risk of death or serious injury.

With regard to point (d), interactions between ships and at-mospheric electricity have been observed and recorded for thou-sands of years. However, Thornton et al. [6] analysed 12 years of high-resolution global lightning stroke data from the World-wide Lightning Location Network (WWLLN) and found that lightning strike density is enhanced by up to a factor of two di-rectly over shipping lanes in the northeastern Indian Ocean and the South China Sea in comparison to adjacent areas with similar climatological characteristics. These authors hypothesised that the increase is due to emissions of aerosol particles.

In a recent study, Petersen [8] has expressed a contrasting hy-pothesis regarding the lightning enhancement found by Thornton et al. [6], namely that tall objects in high electric field environ-ments are known to initiate lightning. Hence, the alternate expla-nation for the lightning enhancement is that tall, well-grounded ships may be facilitating lightning production – particularly in storms that are near the tipping point between remaining “electri-fied shower clouds” and becoming thunderstorms.

Regardless of which hypothesis is more likely, the fact remains that lightning activity along shipping lanes is double what it would be for similar but relatively unnavigated waters. This fact implies that the risk of a lightning incident is double what it would normally be over water.

The main points and research addressed in the remainder of this paper are (i) the author’s quantitative lightning risk analysis for vessels and their personnel, using a tailored approach that combines traditional methods presented in standards as well as newer methods, (ii) a review of the type and extent of damage that occurs on vessels as a result of lightning strikes, and (iii) a step-by-step outline of a systematic lightning protection strategy to mitigate the risks and losses due to lightning.

2. Lightning Risk Analysis

The aim of this section of the paper is to present some calcula-tions of the risks posed by lightning when vessels encounter or operate in thunderstorms. Lightning is an external electrical haz-ard that adds to the common electrical accidents and incidents that can occur due to internal electrical faults on vessels. A simple event tree diagram of the hazards posed by lightning on vessels is shown in Figure 1. On this basis, the paper emphasises the need to assess and mitigate the risks associated with lightning strikes to ensure the safety of vessels and their occupants.

Figure 1:

Generic event tree diagram for lightning hazards on vessels. This diagram assumes personnel are not so exposed that they would be struck directly, the consequence of which may be death

Much of the equipment in vessels at sea is critical. Damage to this equipment due to lightning strikes can lead to the loss of electrical power supply, propulsion, and navigation control. The paper underscores the severe risks posed by fire hazards result-ing from arcing and sparking events on vessels, which can be triggered by lightning strikes. Navigation accidents can occur, since steering, navigation radar, fire pumps, and engine controls are highly critical equipment that are sensitive to electrical surges. Lightning is a common cause of such effects, hence the im-portance of dealing with it appropriately. Critical equipment onboard cannot usually be substituted, so once it catches fire, total loss of vessel control and an accident becomes inevitable.

Despite the importance of assessing lightning risks to vessels, there is limited availability of published literature on lightning risks specific to vessels. Hence, more research and analysis are needed in this area to better understand and address lightning risks in the maritime industry. However, Nicolopoulou et al. [9] have undertaken such a study. These authors calculated the ex-pected number of lightning strikes on three ship models with comparative application of various lightning attachment models and stroke current distributions. They found that, in coastal areas, a ship is expected to be struck by lightning approximately every two years. These authors also proposed a new method for light-ning protection of critical masts on ships that combines a shield-ing analysis procedure and a statistical lightning interception model. This aspect will be addressed in Section 4 of the paper.

The scarcity of literature on this topic emphasises the im-portance of assessing lightning risks to vessels due to the critical nature of onboard equipment and the potential catastrophic con-sequences of lightning-induced accidents. Proper assessment and mitigation of lightning risks are crucial for maintaining operation-al efficiency and ensuring the safety of personnel onboard ves-sels.

2.1 Study Assumptions

The lightning risk study needs to encompass a practical range of vessel sizes (and heights) and allow for the wide range of lightning activity encountered around the world.

Table 1 shows the range of vessel dimensions included in the lightning risk study. Separate risk calculations will be made for each vessel category in Table 1.

Table 1:

Range of vessel types and sizes assumed for the lightning risk calculations. The parameters L, W and H are the vessel’s overall length, beam width and height above the waterline (“freeboard”) to the highest point, respectively. The latter (H) is a key parameter for estimating lightning strike risk

Lightning activity is typically expressed as a “ground flash density” (GFD), in units of ground flashes per square km per year. GFD over land ranges from about 0.5 to more than 30 flashes/km²/yr. A NASA map of world lightning activity is shown in Figure 2.

Figure 2:

World Lightning Map produced by NASA's Lightning Imaging Sensor on the Tropical Rainfall Measuring Mission satellite between 1995 and 2002. The map shows the average ground flash density in units of lightning flashes to Earth per square kilometre per year.

Mackerras et al. [1] estimated that the ratio of the mean global land-to-ocean total flash density is about 3.4. Hence, a reasonable range of GFD for lightning over bodies of water or oceans is 0.1 to 10 flashes/km²/yr. This range will be used for the lightning risk analysis. Note that lightning activity in littoral (shoreline / coastal) areas is likely to be somewhere between the land and ocean values.

2.3 Lightning Incidence Calculations

Tables 2 and 3 present the results of the lightning incidence calculations made for the different vessels listed in Table 1, using the simple 3H, EGM and Rizk methods. As can be seen, different methods give different results, with no clear trend to choose just one method. Given these variations, the mean value of all three methods was used to progress to the probability calculations.

Table 2:

Lightning incidence calculations for the vessels listed in Table 1, using the “simple 3H”, “electrogeometric model” and Rizk analytical formulae. These calculations assume the exposure time is 365 days per year. The range of the results corresponds to the assumed GFD range, i.e., 0.1 – 10 flashes/km2/yr

Table 3:

Mean time interval between lightning flashes (MTBF) to the vessels, using the mean lightning incidence from the three methods investigated. Once again, the range of the results corresponds to the assumed GFD range, i.e., 0.1 – 10 flashes/km2/yr, to which each vessel may be exposed.

Table 3 shows the mean time between flashes to the range of vessel categories included in the risk analysis. Generally speak-ing, direct strikes to vessels can be expected once every year to few years in higher lightning areas, and once every hundred to several years for low lightning areas.

2.4 Probability Calculations

The previous section quantified the likelihood of lightning flashes to vessels. Attention now turns to the consequences of those strikes. Broadly speaking, lightning can cause equipment / operational / economic damage and can even start fires, particu-larly but not limited to vessels carrying flammable or explosive materials. These aspects are addressed in Section 3 of the paper. In this and the next sub-section of the paper, the focus will be on personnel hazards due to lightning strikes.

The probability of occurrence of a potentially hazardous light-ning incident involving a person on a vessel is given by:

$\[ {\mathrm{P}}_{\mathrm{i}\mathrm{n}\mathrm{c}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{ }}={\mathrm{P}}_{\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{k}\mathrm{e}\mathrm{ }}\times \mathrm{ }\mathrm{P}\mathrm{e}\mathrm{x}\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{e} \]$

(5)

where P_incident is the probability of an incident that may involve a person during a thunderstorm, P_strike is the probability of a direct lightning flash to the vessel and P_exposure is the probability that someone will be in an exposed or dangerous position at the in-stant of a lightning flash to the area.

The values for P_strike for the vessel have already been obtained (see Table 2). These values are based on 24 hours-per-day, 365 days-per-year collective exposure to lightning.

The values for P_exposure are based on behavioural patterns that may vary during a 24-hour period or from day to day. The worst-case scenario would be personnel exposure on a vessel for 24 hours-per-day, 365 days-per-year, but this extreme may not be realistic. Hence, 50% of this exposure time is assumed for the probability calculations.

The probability calculations are shown in Table 4.

Table 4:

Estimated probability of a personnel incident occurring on a vessel under the assumptions stated and for the assumed GFD range of 0.1 – 10 flashes/km2/yr.

2.5 Personnel Risks on Vessels

According to IEC 62305-2 [10], the risk component related to injury of human beings, RA, is given by:

$\[ {\mathrm{R}}_{\mathrm{A}}={\mathrm{N}}_{\mathrm{D}}\times {\mathrm{P}}_{\mathrm{A}}\times {\mathrm{L}}_{\mathrm{A}} \]$

(6)

where:

• • ND is the flash incidence given by Equation (2),
• • PA is the probability that the lightning flash will cause an injury to human beings via one of the mechanisms outlined in the literature, e.g., Cooper et al. [18], typi-cally some form of electric shock, or a trauma incident such as acoustic shock wave, e.g., Gluncic et al. [19], and
• • LA is the “consequent loss” (of human life) or serious injury, which depends on the proportion of people pre-sent and at risk, and the amount of exposure time.

If an ALARP risk approach is taken, then the maximum prob-ability for P_A must be used, i.e., 1.0. On the other hand, lower values can be used if justified, e.g., where extensive lightning protection measures are provided. In this analysis, it will be as-sumed that no lightning protection measures are in place.

With regard to N_D and L_A, the probability calculations in Table 4 have already taken into account the former and part of the latter (via the exposure time). However, L_A is more rigorously defined as follows:

$\[ {\mathrm{L}}_{\mathrm{A}}={\mathrm{r}}_{\mathrm{t}}\times L_T\times \left({\mathrm{t}}_{\mathrm{z}}/8760\right) \]$

(7)

where r_t is a “loss reduction factor” that depends on the type of soil or flooring, LT is the is the typical mean percentage of per-sons injured by a lightning flash, and t_z is the time in hours per year for which the persons are present in the area of interest.

Standards [10][11] suggest a value of 0.01 for L_Tif the person is in some form of structure or shelter when the incident occurs. For people on the deck of a vessel, this scenario is not applicable. Such people may also be adjacent to a structural element of the vessel and hence at risk of a side-flash. The exact value to use in this case is difficult to estimate as the international literature, ob-viously, tends to focus on deaths and injuries rather than, for example, “near misses” with no injuries. However, according to Ritenour et al. [20], up to 80% of people involved in a lightning incident receive some form of injury. Hence, a conservative value of 0.8 will be used.

With regard to r_t, Table C.3 from IEC 62305-2 [10] is used, where the value of r_t ranges from 10^–2 for the most conductive surfaces to 10^–5 for materials with a high voltage withstand. As personnel on vessels will often be standing on metallic surfaces, a value of 10^–2 must be used.

Individual (single person) risks were computed using the guidelines above. These risks are shown in Table 5, noting that they are all “per person” risks. Hence, for N people exposed to lightning, the risks in Table 5 should be multiplied by N.

Table 5:

Outcome of lightning risk calculations (annualised) that are applicable to persons aboard vessels. The values shown are applicable to a single person at risk, i.e., they are “per person” risks. As before, these risks were calculated for an assumed GFD range of 0.1 – 10 flashes/km2/yr

Standards [10][11] typically nominate 1.0 x 10^–5 yr^-1 as a “tol-erable” lightning risk for loss of human life or serious injury. Comparing this value with the calculations summarised in Table 5, the personnel risk is seen to be intolerable for the majority of vessel types across the range of typical lightning activity en-countered over bodies of water.

3. Physical Damage

4. Mitigation Solutions

Earlier sections of the paper have described the risks posed by lightning to personnel and marine vessels. In most cases, the lightning risks are intolerable, so mitigation measures need to be put into place.

This section of the paper briefly describes the historical devel-opments behind the protection of marine vessels, provides a ge-neric, comprehensive, four-step approach to lightning protection, and then focuses on the specific lightning protection measures needed on marine vessels to mitigate the risks to tolerable levels.

4.2 Comprehensive Approach to Lightning Mitigation

This section of the paper describes proven measures to reduce asset and personnel risks due to lightning strikes. Note that it is not possible to completely eliminate the risk of loss due to light-ning and this is why there are internationally accepted “tolerable risk” values. However, a large reduction in the risk can be achieved via a systematic and comprehensive approach to light-ning protection that considers the major damage mechanisms and potential losses due to lightning.

At an overview level, the conceptual graphic shown in Figure 3 summarises the comprehensive approach that needs to be taken to minimise losses due to lightning. This approach is comprised of four key steps:

• I. Protect critical assets against direct lightning strikes.
• II. Protect electrical and electronic equipment against surges and transients.
• III. Provide a low impedance reference earth / ground and bond all conductive elements to minimise voltage differ-ences.
• IV. Protect people.

Figure 3:

Conceptual graphic summarising the four key steps that need to be implemented to reduce the risk of losses due to lightning to tolerable levels (reprinted with permission from Lightning Protection International Pty Ltd).

This approach is applicable across all industry sectors, e.g., ag-riculture, aviation, marine, commercial, industrial, and recreation-al buildings and facilities, cultural, defence, mining, petrochemi-cal / oil & gas, power & renewables, communications, transporta-tion and utilities. It is applicable to any system, structure (includ-ing vessels), site or facility in which maintaining operational efficiency, minimising losses, and keeping personnel safe is nec-essary or important.

The remainder of this section provides more details on each step as it applies to vessels.

4.2.1 Step I – Protection Against Direct Strikes

Step I involves the capture of lightning strikes with “air termi-nals” (sometimes called “lightning rods” or “air terminations”). Air terminals must be strategically positioned at high-risk points to minimise the possibility of lightning bypass, a process some-times called “shielding”. Some of the key points in Step I include:

• • Effective shielding depends on the protection area provid-ed by each air terminal and the positioning of the air ter-minal(s) to achieve the desired “interception efficiency” or “lightning protection level” (LPL).
• • Both of these aspects must be taken into account by the “lightning protection design method” that is used. In gen-eral, the tallest and most exposed points on a vessel are the most vulnerable to a direct strike, so the correct instal-lation of air terminals near or on these vulnerable points is a key element of Step I.
• • The lightning current must then be carried down to ground in a safe manner with “downconductors”, away from sensitive equipment and personnel. Hence, the pre-vention of dangerous sparking between the downconduc-tors and internal conductive components (conduits, pipes, equipment chassis, incoming and outgoing conductors, etc.) is essential. Dangerous sparking between different parts can be avoided by means of “equipotential bonding” or electrical insulation between the parts.
• • Hence, there are generally two application methods and, consequently, two types of downconductors used, name-ly:
  • (i) Protection by “equipotential bonding”, utilising bare downconductors or the “natural components” of the vessel, and
  • (ii) Protection by “isolation”, either utilising insulating materials (as in the case of insulated downconduc-tors) or a suitable air gap or “separation distance”. Equation (8) is a simple way of calculating the sep-aration distance, s in metres [10][11]:
  • where l = longest length of the downconductor path to ground with no equipotential bonding point in metres, k_i is 0.08, 0.06 or 0.04 for LPL I, II or III/IV respectively, k_c is 1.0, 0.66 and 0.44 for 1, 2, and ≥ 3 downconductors, and k_m is 1.0 or 0.5 for air or concrete / bricks / wood respectively.

$\[ s=\frac{k_i{k}_c}{k_m}l \]$

(8)

4.2.2 Step II – Protection Against Surges and Transients

Step II is required to mitigate overvoltages, surges and transi-ents due to lightning, as discussed in Section 3.3. Such overvolt-ages can damage or destroy primary and secondary electrical / electronic equipment within a vessel. This so-called “surge pro-tection” is achieved via the use of “surge protection devices” (SPDs).

There are many aspects to consider with the application of SPDs. A brief summary of the main aspects of this vast topic includes the following key points:

• • The aim is to limit residual voltages to within the immuni-ty level (or withstand voltage) of the internal equipment.
• • Surge protection technologies can generally be classified into two categories, namely “shunt SPDs” and “surge fil-ters”, which are parallel and series protection devices re-spectively.
• • Electrical line being protected – power vs signal / data / communications will determine the type of SPD required.
• • Primary (or “point-of entry”) vs secondary protection – the former needs to be more robust to deal with larger overvoltages and the latter needs to cater for sensitive equipment “down the line”.
• • Some of the parameters that need to be considered when selecting SPDs include:
  • o Maximum Continuous Operating Voltage (MCOV) – a major safety consideration, e.g., prevention of fires.
  • o Clamping voltage – to protect the downstream equipment.
  • o Surge rating or line current – ability of the SPD to handle surge current (typically given in kA).
  • o Protection modes – Line-Neutral (L-N), Line-Earth (L-E), Neutral-Earth (N-E), Line-Line (L-L).
  • o Indication and life – older SPDs use indicator LEDs, whereas “smart SPDs” based on Bluetooth and oth-er wireless technologies are now revolutionising this aspect of SPD maintenance and service [27].
• • Transients induced onto data, communication and signal lines can easily damage and destroy sensitive terminal equipment and hence lead to down-time. Protection of communications equipment requires the same concepts as those noted above, such as:
  • o SPDs for sensitive equipment are typically multi-stage or series-connected devices with much lower operating currents and voltages. These SPDs are in-stalled at the point of entry to the structure or at the equipment termination itself.
  • o Internal wiring that extends more than 10-15 m should also be protected. Twisting or shielding of cables provides some protection. However, this practice should not be regarded as sufficient for the sensitive interfaces that characterise modern com-munication devices.

4.2.3 Step III – Provision of Good Earthing and Bonding

Step III is fundamental to the overall lightning protection scheme as it minimises “earth potential rise” (EPR) effects and hence helps to prevent fires and other damage due to sparking. However, this step is where vessels differ greatly from most other applications. Underneath the vessel exists a body of water (fresh or saline) that can act as a very good “earthing system” without the huge variation seen in ground- or soil-based earthing outcomes as a result of the large variation in soil resistivity.

Therefore, the key to good earthing on vessels is to ensure that all lightning downconductor paths are directly bonded to metallic items in contact with the water at all times. This task is relatively easy to achieve in vessels with metallic hulls, whereas vessels with non-metallic hulls require the installation of one or more dedicated “earth plates” under the hull^‡.

Rigorous equipotential bonding is also necessary, particularly for a non-isolated lightning protection system where the vessel is electrified by the lightning strike. Bonding is primarily a voltage consideration and is accomplished successfully by connecting all conductive elements together to a single point of reference in a “star network” arrangement. Bonding should be carried out using suitable conductor sizes (typically 35 – 50 mm²) and connections. The bonding path must be kept as short as possible so that a damaging voltage differential does not exist between the end of the bonding conductor and other conductive components within the vessel.

4.2.4 Step IV – Personal Protection

Step IV is the important task of protecting people (crew, pas-sengers, etc. hereafter called “personnel”) against lightning strikes. Much has been written about the effects of lightning on human beings, e.g., [18][28]. There are four main electrical mechanisms associated with lightning strikes that make cause injury or death to human beings, namely a:

• • Direct strike to the person.
• • Side flash from an adjacent structure, e.g., while shelter-ing beneath or near a (tall) tree.
• • Touch voltage, i.e., contact with a conductor that has risen to a dangerous voltage that drives a dangerous or fatal current through the person’s body to ground.
• • Step voltage, i.e., an indirect cloud-to-ground strike that causes a large voltage in the soil or, in some cases, sur-face arcing along the ground which can create a voltage difference across the human body via the feet and legs.

The effects of these electrical mechanisms include burns to the skin, damage to various bodily organs and systems, uncon-sciousness, and death.

The lightning effects outlined above can be reduced dramatical-ly via the installation of a properly designed lightning protection system by following the four steps in this paper. However, it is very difficult to protect personnel from all lightning hazards. Hence, ideally, no personnel should be out on deck / in the open during a thunderstorm, i.e., administrative / procedural controls should be used. According to AS 1768 [11], to the extent con-sistent with safe handling and navigation of the vessel during a lightning storm, personnel should:

• • Remain inside a closed vessel and avoid contact with me-tallic items.
• • Stay as far as practicable from any items forming part of a downconductor path for the lightning current.
• • Not be in the water, or dangle arms or legs in the water.

Even with a lightning protection system in place, avoiding con-tact with metallic items is very important. As shown by Nicolopoulou et al. [29] with electromagnetic simulations of lightning strikes to a ship, touch voltages as high as 19 kV were found in the bridge area of the ship. These authors explained that the dissipation of the lightning current on the ship’s surface caus-es voltage rise of the hull which acts as the ground reference level for the ship’s electric grid and development of dangerous step and touch voltages.

There are also three main non-electrical mechanisms that can cause serious human injuries, namely:

• • Acoustic pressure wave(s) from nearby thunder causing acoustic injuries, e.g., ruptured ear membranes or tinnitus.
• • Radiation from the lightning channel causing eye damage, e.g., temporary blindness, vision damage, cataracts, etc.
• • Flying or falling debris from structures struck by light-ning.

Once again, these hazards are avoided by remaining inside the closed vessel, although appropriate PPE can also lower the risks.

In the next section, some practical solutions are presented for addressing the risks and hazards outlined in the four steps above.

4.3 Practical Solutions

In summary, the comprehensive, four-step lightning protection plan recommended in this paper requires consideration and im-plementation of (i) direct-strike protection, (iii) surge protection, and (iii) earthing and bonding, with (iv) personal protection as an additional factor in any situations where personnel are at risk.

Within Step (i), there are two aspects to capturing lightning strikes reliably, namely (a) the methodology used for positioning air terminals, and (b) the air terminal “hardware” used. Both as-pects require further explanation before a practical solution can be presented for direct-strike protection – see Subsections 4.3.1 and 4.3.2. Finally, Subsection 4.3.3 presents a proven, effective solu-tion for protecting vessels against lightning strikes.

4.3.1 Methodology

The “rolling sphere method” (RSM) is commonly used for po-sitioning air terminals on structures [10][11]. It implements the simple “electrogeometric model” (EGM). To apply the RSM to vessels, an imaginary sphere with a radius equal to the striking distance calculated from the EGM, typically 45 metres, is rolled over the vessel in 3D. All surface contact points are deemed to require protection, whilst the unaffected surfaces and volumes are deemed to be protected.

The advantage of the RSM is that it is conceptually simple, even for application to vessels with complicated shapes. Howev-er, since it is a simplification of the physical process of lightning attachment to a vessel, it has some limitations. The main limita-tion is that it assigns an equal upward leader initiation and light-ning attachment probability to all contact points on the vessel, i.e., no account is taken of the role of electric field enhancement in upward leader initiation. Furthermore, when the RSM is applied to a vessel of height greater than the selected sphere radius, the sphere touches all parts of the vertical sides of the vessel struc-tures above a height equal to the sphere radius.

The limitations of the RSM have led many researchers to de-velop improved lightning attachment models for air terminal placement, e.g.,

• • “Leader progression model” of Dellera & Garbagnati [30],
• • “Collection volume method” of Eriksson [12][13], later expanded to extended structures as described in D’Alessandro et al. [31][32],
• • “Simplified leader inception model” of Becerra & Cooray [33][34], and
• • “Leader inception theory” of Rizk [35]-[37].

All of these models have “leader propagation” at their heart, i.e., allow for the propagation of upward and downward leaders.

Referring specifically to vessels, Cvjetković et al. [38] state that most vessels are “non-conventional” structures, particularly under the dynamic conditions of rolling, pitching and swinging. They assert that existing “standardised methods are insufficient and do not provide the required level of safety”. These authors also proposed the use of a lightning strike warning system on board the vessel, where the crew could be alerted in real time. This suggestion fits in well with mitigation of the personnel risks discussed in Section 2 of this paper.

Hossam-Eldin & Omran [39] presented a technique to use the Collection Volume Method (CVM) for the placement of conven-tional or non-conventional lightning protection systems on ves-sels. These authors made calculations that included the vessel height and dimensions, radius of curvature, location, risk factors, and lightning parameters. They applied the method to a range of vessels, i.e., medium-sized war, cargo, destroyer, and aircraft carrier. They showed the method to be a very efficient means for lightning protection of vessels.

Hossam-Eldin & Abdalla [40] also published a new concept for lightning protection of vessels, using a “leader potential con-cept”. This method is based on a publication by Mazur & Ruhn-ke [41] that suggested striking distance can be estimated from the potential of the downward lightning leader. This transfer of con-cept in [40] resulted in lightning protection systems that were the least conservative (or most efficient) when compared to the RSM and CVM when applied to a warship, frigate, destroyer, aircraft carrier and cargo vessel. However, their analysis did not consider upward leaders. Out of the three methods they compared, only CVM takes the upward leader into account.

Fortunately, the work of Rizk [35]-[37] solves the above defi-ciency. Rizk’s “leader inception theory” (LIT) uses the concept of space potential and considers both the downward and upward leaders in the lightning attachment process. The LIT is universal-ly applicable, i.e., can deal with lightning protection problems in power systems, ordinary buildings, vessels, etc.

In summary, improved lightning attachment models have been applied successfully to direct-strike protection of vessels. Hence, there is no need to rely solely upon the RSM.

4.3.2 Hardware

During a thunderstorm, the electric field between the thunder-cloud and the vessel (and surrounding water) is at highly elevated levels. Under such conditions, “corona discharge” emanates from many points on a vessel, as well as from any air terminals in-stalled for lightning protection, during the so-called “pre-strike phase of the storm. Historically, this effect was seen commonly on sailing ships and was called “St Elmo’s Fire” [42].

In terms of lightning protection, corona discharge results in the development of a space charge “volume” or “cloud” above the object(s) [43]. Sharp-tipped air terminals are well known to pro-duce substantially more corona space charge than blunt-tipped air terminals [44][45]. There is now significant theoretical and ex-perimental evidence that space charge accumulation around the top of an air terminal has a detrimental effect on its ability to initi-ate and sustain an upward leader [46]-[49][36][50]-[56]. The outcome of this space charge effect is that lightning capture is much less reliable because it affects the ability of the air terminal to launch the continuous, uninhibited upward leader that is need-ed for reliable interception of the downward lightning leader.

Hence, the most basic fact about lightning protection is that is far better to install optimised air terminals at the most likely strike point(s) on the vessel. The correct placement of air terminals is achieved with a suitable lightning attachment model as described in Subsection 4.3.1, preferably a modern leader propagation model that can account for the variables and parameters involved in the lightning attachment process e.g., D’Alessandro et al. [31]-[32][45], Rizk [35]-[37].

The most important optimised parameter is the geometry of the air terminal. This aspect was first studied systematically by Moore [46] and more recently by other researchers [55][56]. The optimum air terminal is one that is corona minimising during the pre-strike phase of a thunderstorm but, upon the initial descent of the downward leader, commences the corona-streamer-leader process in a dynamic response that leads to a continuously prop-agating upward leader and ultimate interception of the downward leader. Calculations of corona onset for given air terminal shapes rely on basic gas discharge physics [57] and calculations of the optimum air terminal geometry for practical installations are also achievable [58][59].

All of the above concepts and research are described in more detail in D’Alessandro [60].

4.3.3 Proven Lightning Protection Approach

A comprehensive lightning protection approach for vessels can be formulated from all the principles outlined in earlier sections of this paper. The generic concept for this approach is illustrated in Figure 4.

Figure 4:

Conceptual drawing showing the proven, generic approach to protecting vessels against lightning strikes. Key: ---- HVSC Plus insulated lightning downconductor cable,    Surge protection devices (SPDs) on equipment, ---- Surge proteon bonds to earth, ---- Bonding of vessel conductive elements to earth, ---- Vessel earthing.

Capture of lightning strikes at a prominent location (deter-mined by the design methodology) is with one or more corona minimising air terminals, e.g., “Guardian CAT” or “Guardian Plus” [45]. The reasons for utilising this technology were ex-plained in Section 4.3.2. The remainder of the direct-strike sys-tem implements “protection by isolation”. This approach is a very effective for protecting vessels, i.e., discharge the lightning cur-rent into the water without electrification of any parts of the ves-sel.

The use of at least 2 metres of FRP mast together with an insu-lated lightning downconductor cable with an impulse withstand voltage of at least 500 kV [61] provides the necessary insulation and prevention of electrification of the vessel. This “HVSC Plus” cable minimises the risk of side flashes, a common problem with downconductors carrying fast rising lightning currents. Such incidents must be prevented on vessels because the arc flash can lead to a fire or blast in the nearby area, causing fatalities and losses.

A “lightning strike recorder” (LSR) may also be installed if in-formation about the lightning strikes captured by the system is required. This information may include parameters such as date and time, size of strike (peak current), etc. and may be accessible remotely for convenience and keeping maintenance costs to a minimum.

Figure 5 is a photo of a typical direct-strike installation on a vessel.

Figure 5:

Typical direct-strike installation on a vessel that uses a “protection by isolation” approach

(b) Surge Protection

Surge protection must be applied to all valuable and critical equipment on vessels. Such equipment has an incoming power line as well as data / signal / control lines, depending on their function. All of these lines may carry surges into the equipment and damage or destroy it, hence they all need appropriately se-lected surge protection devices (SPDs). For power lines, the choices depend on single- vs three-phase SPDs, shunt vs series, type of electrical system, etc.

Hence, the surge protection requirements of every vessel are different. They depend upon the size, function and nature of the vessel. A common example of power line protection is the DIN Rail-mounted 3DR100KA-385-NE100 surge protection setup. The same components are also available in a compact “power protection module”, e.g., 3PPM100kA-385-NE100-AIMCB, allowing easier installation in many cases.

“Smart SPDs” are now available that allow monitoring of key physical parameters and provide easier indication methods for when SPDs need to be replaced. Further details on surge protec-tion generally, and smart SPDs particularly, can be found in D’Alessandro [27].

Figure 6 is a photo of a typical surge protection installation for power lines.

Figure 6:

Typical surge protection installation for power lines

(c) Earthing and Bonding

The earthing and bonding arrangements of a vessel will also vary per the size, function, and nature of the vessel. Earthing is the final point of contact where the lightning strike safely dissi-pates into the water, whilst bonding conductors and connections ensure that lightning currents are controlled and dissipated into the water. Bonding conductors should have sufficient cross-section areas to ensure the full or partial lightning currents can be carried safely. The cross-sectional area typically varies from around 16 mm² to 50 mm² [11][62].

Figure 4 shows an earth plate (in green) under a vessel as a standard method of earthing for vessels with non-metallic vessel hulls. For vessels with metallic hulls, the hull provides a very good earth plane. It is important to ensure direct bonding ar-rangements for all equipment, SPDs and conductive elements occur to a common point on the hull.

(d) Personal Protection

In Section 2.5, it was concluded that personnel risk is seen to be intolerable for the majority of vessel types across the range of typical lightning activity encountered over bodies of water. The best risk reduction measure is elimination, i.e., personnel on ves-sels are not outside in thunderstorms.

The problem with elimination is the decision-making around when shelter should be sought. This decision is best made with a “lightning warning system” (LWS), as also pointed out by Cvjetković et al. [38]. A LWS that detects all phases of a thun-derstorm and provides a warning based on the magnitude of the atmospheric electric field is highly recommended.

It is recognised that elimination of risk may not always be pos-sible. If this is the case, personnel should be wearing full PPE, which includes:

• (i) Eye wear that is close-fitting with a wrap-around design and that blocks at least 99% of the UV radiation spectrum.
• (ii) Ear plugs or earmuffs that provide a noise reduction of at least 30 dB, but preferably up to 120 dB.
• (iii) Safety boots with soles that have a good withstand volt-age, nominally at least 5 kV when wet and preferably up to 20 kV under a dry test.

Furthermore, contact with any conductive elements of the ves-sel must be avoided.

5. Conclusions

This paper has investigated the risk of asset and operational losses, as well as loss of human life, on vessels due to lightning storms. Such losses have even more serious consequences than land-based losses due to the remote locations of vessels, com-promising safety onboard.

A detailed quantitative analysis of the risks posed by lightning to vessels and their personnel was carried out. It was found that the mean time between flashes to the range of vessel categories investigated is once every year to few years in higher lightning areas, and once every hundred to several years for low lightning areas. Hence, the risk of a lightning strike to a vessel cannot be ignored. Furthermore, the risk to personnel was to be intolerable for the majority of vessel types across the range of typical light-ning activity encountered over bodies of water.

The paper then described the physical damage that is known to occur when lightning strikes vessels directly or indirectly and a comprehensive lightning mitigation approach was presented. This four-step approach is comprised of direct-strike protection, surge protection, earthing and bonding, and personal protection.

A generic and practical lightning protection solution was pro-vided for vessels using some of the outcomes of the latest inter-national research on lightning protection. This solution is appli-cable to all types of vessels, i.e., it has no dependence on the material of the hull, size of the vessel, type of electrical systems on board, etc. However, the specific details and material require-ments will depend on the above variables.

With the increasing sensitivity of equipment and the im-portance of human life, lightning protection of vessels is essen-tial. With appropriate measures in place, safer and more efficient vessel operations can be ensured.

Author Contributions

F. D’Alessandro: Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization.

References

• Mackerras, D., Darveniza, M., Orville. R.E., Williams, E.R. and Goodman, S.J., “Global lightning: Total, cloud and ground flash estimates,” Journal of Geophysical Research, vol. 103, no. D16, pp. 19791-19809, 1998. [https://doi.org/10.1029/98JD01461]
• Romps, D.M., Seeley, J.T., Vollaro, D. and Molinaria, J., “Projected increase in lightning strikes in the United States due to global warming”, Science, vol. 346, no. 6211, pp. 851-854, 2014. [https://doi.org/10.1126/science.1259100]
• Wilson, J.G. and Cummins, K.L., “Thunderstorm and fair-weather quasi-static electric fields over land and ocean,” Atmospheric Research, vol. 257, 105618, 2021. [https://doi.org/10.1016/j.atmosres.2021.105618]
• Asfur, M., Price, C., Silverman, J. and Wishkerman, A., “Why is lightning more intense over the oceans?,” Journal of Atmospheric and Solar-Terrestrial Physics, vol. 202, 105259, 2020. [https://doi.org/10.1016/j.jastp.2020.105259]
• Rakov, V. A. and Uman, M. A., Lightning: Physics and Effects, London: Cambridge University Press, 2007.
• Thornton, J.A., Virts, K.S., Holzworth, R.H. and Mitchell, T.P., “Lightning enhancement over major oceanic shipping lanes,” Geophysical Research Letters, vol. 44, no. 17, pp. 9102-9111, 2017. [https://doi.org/10.1002/2017GL074982]
• Holzworth, R.H., McCarthy, M.P., Brundell, J.B. and Ja-cobson, A.R., “Global distribution of superbolts”, Journal of Geophysical Research: Atmospheres, vol. 124, pp. 9996-10005, 2019. [https://doi.org/10.1029/2019JD030975]
• Petersen, M.J., “Interactions between lightning and ship traffic,” Earth and Space Science, vol. 10, no. 11, p. e2023EA002926, 2023. [https://doi.org/10.1029/2023EA002926]
• Nicolopoulou, E.P., Gonos, I.F. and Stathopulos, I.A., “Lightning shielding analysis on ships,” IEEE Transactions on Transportation Electrification, vol. 3, no. 3, pp. 779-791, 2017. [https://doi.org/10.1109/TTE.2017.2671520]
• IEC 62305-2, Protection Against Lightning – Part 2: Risk Management”, 2nd Edition Geneva, Switzerland, 2010.
• AS 1768, Lightning Protection, Standards Australia Tech-nical Committee EL-024, Sydney, Australia, 2021.
• Eriksson, A.J., “The incidence of lightning strikes to power lines,” IEEE Transactions on Power Delivery, vol. 2, pp. 859-870, 1987. [https://doi.org/10.1109/TPWRD.1987.4308191]
• Eriksson, A.J., “An improved electrogeometric model for transmission line shielding analysis,” IEEE Transactions Power Delivery, vol. 2, pp. 871-886, 1987. [https://doi.org/10.1109/TPWRD.1987.4308192]
• Rizk, F.A.M., “Modeling of transmission line exposure to direct lightning strokes,” IEEE Transactions on Power De-livery, vol. 5, pp. 1983-1997, 1990. [https://doi.org/10.1109/61.103694]
• Rizk, F.A.M., “Modeling of lightning incidence to tall struc-tures. Part I: Theory”, IEEE Transactions on Power Deliv-ery, vol. 9, pp. 162-171, 1994. [https://doi.org/10.1109/61.277673]
• Rizk, F.A.M., “Modeling of lightning incidence to tall struc-tures. Part II: Application”, IEEE Transactions Power De-livery, vol. 9, pp. 172-193, 1994. [https://doi.org/10.1109/61.277690]
• D’Alessandro, F. and Petrov, N.I., 2006, “Field study on the interception efficiency of lightning protection systems and comparison with models,” Proceedings of the Royal Society A, vol. 462, pp. 1365-1386, 2006. [https://doi.org/10.1098/rspa.2005.1625]
• Cooper, M.A., Holle, R.L. and Andrews, C., “Distribution of Lightning Injury Mechanisms”, 2010 30th International Conference on Lightning Protection, Tucson, USA, 2010. [https://doi.org/10.1109/ICLP.2010.7845948]
• Gluncic, I., Roje, Z. and Gluncic, V., “Ear injuries caused by lightning: report of 18 cases,” Journal of Laryngology & Otology, vol. 115, pp. 4-8, 2001. [https://doi.org/10.1258/0022215011906858]
• Ritenour, A.E., Morton, M.J., McManus, J.G., Barillo, D.J. and Cancio, L.C., “Lightning injury: a review”, Burns, vol. 34, no. 5, pp. 585–594, 2008. [https://doi.org/10.1016/j.burns.2007.11.006]
• Salway, R., “Inland Hull, Fishing Vessel & Yacht Market – An Underwriting View”, IUMI EYE Newsletter, p. 13, 2019.
• AGCS, “Loss log: Lightning,” Global Risk Dialogue Sum-mer/Autumn 2019, 2019.
• Nicolopoulou, E.P., Gonos, I.F. and Stathopulos, I.A., “Lightning induced overvoltages in the electric network of a ship,” IEEE International Conference on High Voltage En-gineering and Application (ICHVE), Athens, Greece, 2018. [https://doi.org/10.1109/ICHVE.2018.8642224]
• Bernstein, T. and Reynolds, T.S., “Protecting the royal navy from lightning – William Snow Harris and his struggle with the british admiralty for fixed lightning conductors,” IEEE Transactions on Education, vol. 21, no. 1, pp. 7-14, 1978. [https://doi.org/10.1109/TE.1978.4321177]
• Harris, W.S., Protection of Ships from Lightning: Accord-ing to Principles Established, Sleeper & Rogers, Boston, 1848.
• Uman, M.A., The Lightning Discharge, Academic Press, Florida, 1987.
• D’Alessandro, F., “Smart SPDs: The future of surge pro-tection,”, Down To Earth Conference (DTEC 2022), Hunter Valley, Australia, 8-9 June 2022.
• Berger, G., “Lightning-caused accidents and injuries to humans”, Proceeding from IX International Symposium on Lightning Protection, Invited Lecture, Foz do Iguacu, Brazil, 2007.
• Nicolopoulou, E.P., Oikonomopoulos, A., Gonos, I.F. and Stathopulos, I.A., “Electromagnetic simulation of a lightning strike on a ship”, Proc. 2016 IEEE International Conference on High Voltage Engineering and Application (ICHVE), Chengdu, China, pp. 1-4, 2016. [https://doi.org/10.1109/ICHVE.2016.7800710]
• Dellera, L. and Garbagnati, E., “Lightning stroke simulation by means of the leader progression model. Part I. Descrip-tion of the model and evaluation of exposure of free-standing structures,” IEEE Transactions on Power Delivery, vol. 5, pp. 2009-2022, 1990. [https://doi.org/10.1109/61.103696]
• D'Alessandro, F. and Gumley, J. R., “A ‘Collection Vol-ume Method’ for the placement of air terminals for the pro-tection of structures against lightning”, Journal of Electro-statics, vol. 50, pp. 279-302, 2001. [https://doi.org/10.1016/S0304-3886(00)00044-9]
• D’Alessandro, F., “The use of ‘field intensification factors’ in calculations for lightning protection of structures”, J. Electrostat., vol. 58, pp. 17-43, 2003. [https://doi.org/10.1016/S0304-3886(02)00178-X]
• Becerra, M. & Cooray, V., “A self-consistent upward leader propagation model,” Journal of Physics D: Applied Physics, vol. 39, no. 16, pp. 3708-3715, 2006. [https://doi.org/10.1088/0022-3727/39/16/028]
• Becerra, M. and Cooray, V., “A simplified physical model to determine the lightning upward connecting leader incep-tion,” IEEE Transactions on Power Delivery, vol. 21, pp. 897-908, 2006. [https://doi.org/10.1109/TPWRD.2005.859290]
• Rizk, F.A.M., “Modeling of lightning exposure of buildings and massive structures,” IEEE Transactions on Power De-livery, vol. 24, no. 4, pp. 1987-1998, 2009. [https://doi.org/10.1109/TPWRD.2009.2028759]
• Rizk, F.A.M., “Modeling of lightning exposure of sharp and blunt rods”, IEEE Transactions on Power Delivery, vol. 25, no. 4, pp. 3122-3132, 2010. [https://doi.org/10.1109/TPWRD.2010.2049384]
• Rizk, F.A.M., “Modeling of substation shielding against direct lightning strikes,” IEEE Transactions on Electromag-netic Compatibility, vol. 52, no. 3, pp. 664-675, 2010. [https://doi.org/10.1109/TEMC.2010.2046903]
• Cvjetković, S.J., Kasum, J. and Tokić, T., “Lightning pro-tection on non-convention vessels in dynamic conditions,” International Journal of Engineering Research and Applica-tion, vol. 8, no. 2, pp. 68-74, 2018.
• Hossam-Eldin, A.A. and Omran, E.A.M, “The collection volume method for lightning protectors placement on naval ships”, 2007 Annual Report Conference on Electrical Insu-lation and Dielectric Phenomena (CEIDP 2007), pp. 512-516, 2007. [https://doi.org/10.1109/CEIDP.2007.4451453]
• Hossam-Eldin, A.A. and Abdalla, A.B., “New concept for lightning protection of ships: The leader potential concept method,” 14th International Middle East Power Systems Conference (MEPCON’10), Cairo University, Egypt, PA-PER 185, pp. 392-397, 2010.
• Mazur, V. and Ruhnke, L.H., “Determining leader potential in cloud-to-ground flashes,” Geophysical Research Letters, vol. 29, no. 12, pp. 42-1 to 42-4, 2002. [https://doi.org/10.1029/2001GL014159]
• Golde, R.H. (ed.), Lightning, Volume 1: Physics of Light-ning, Academic Press, London, 1977.
• Soula, S. and Chauzy, S., “Effects of ground coronae dur-ing lightning flashes,” Annales geophysicae. Series B. Ter-restrial and planetary physics, vol. 4, no. B6, pp. 613-624, 1986.
• D’Alessandro, F. & Berger, G., “Laboratory studies of corona emissions from air terminals,”, Journal of Physics D: Applied Physics, vol. 32, pp. 2785–2790, 1999. [https://doi.org/10.1088/0022-3727/32/21/311]
• D’Alessandro, F. and Sequeira, S., “Methodology and im-plementation of leader inception theory for lightning protec-tion of substations,” Proceedings of Down To Earth 2016 Conference, Hunter Valley, NSW, Australia, Day 2, Paper 6, 2016. [https://doi.org/10.1109/DTEC.2016.7731283]
• Moore, C.B., “Improved configurations of lightning rods and air terminals,” Journal of Franklin Institute, vol. 315, pp. 61-85, 1983. [https://doi.org/10.1016/0016-0032(83)90107-2]
• Moore, C.B., Rison, W., Mathis, J. and Aulich, G., “Light-ning rod improvement studies,” Journal of Applied Meteor-ology and Climatology, vol. 39, pp. 593-609, 2000. [https://doi.org/10.1175/1520-0450-39.5.593]
• Moore, C.B., Aulich, G.D. and Rison, W., “Measurement of lightning rod responses to nearby strikes,” Geophysical research letters, vol. 27, pp. 1487-1490, 2000. [https://doi.org/10.1029/1999GL011053]
• Moore, C.B., Aulich, G. and Rison, W., “The case for using blunt-tipped lightning rods as strike receptors,”, Journal of Applied Meteorology and Climatology, vol. 42, pp. 984-993, 2003. [https://doi.org/10.1175/1520-0450(2003)042<0984:TCFUBL>2.0.CO;2]
• Sunaga, Y. and Shindo, T., “Influence of space charge on upward leader initiation”, IEEJ Transactions on Power and Energy, vol. 125, no. 8, pp. 789-796, 2005. [https://doi.org/10.1541/ieejpes.125.789]
• Aleksandrov, N.L., Bazelyan, E.M., and Raizer, Yu.P., “The effect of a corona discharge on a lightning attachment,” Plasma Physics Reports, vol. 31, no. 1, pp. 75-91, 2005. [https://doi.org/10.1134/1.1856709]
• Bazelyan, E.M., Raizer, Yu.P. and Aleksandrov, N.L., “Co-rona initiated from grounded objects under thunderstorm conditions and its influence on lightning attachment”, Plas-ma Sources Science and Technology, vol. 17, pp. 1-17, 2008. [https://doi.org/10.1088/0963-0252/17/2/024015]
• Bazelyan, E.M., Raizer, Yu.P. and Aleksandrov, N.L., “The effect of space charge produced by corona at ground level on lightning attachment to high objects,” Atmospheric Re-search, vol. 153, pp. 74-86, 2015. [https://doi.org/10.1016/j.atmosres.2014.07.018]
• Xie, Y., He, H., He, J. and Wu, C., “The effect of corona discharge on leader initiation in long air gaps,” IEEE Trans-actions on Plasma Science, vol. 42, no. 4, pp. 890-895, 2014. [https://doi.org/10.1109/TPS.2014.2305446]
• Cooray, V., “The influence of lightning conductor radii on the attachment of lightning flashes”, Electric Power Systems Research, vol. 153, pp. 138-143, 2017. [https://doi.org/10.1016/j.epsr.2017.01.002]
• Lysov, N., Temnikov, A., Chernensky, L., Orlov, A., Belo-va, O., Kivshar, T., Kovalev, D., Mirzabekyan, G., Leb-edeva, N. and Voevodin, V., “The physical experimental modelling of the formation processes of upward discharges from grounded objects in the artificial thunderstorm cell’s electric field,” Atmosphere, vol. 13, no. 8, p. 1339, 2022. [https://doi.org/10.3390/atmos13081339]
• Lowke, J.J. and D’Alessandro, F., “Onset corona fields and electrical breakdown criteria,” Journal of Physics D: Ap-plied Physics, vol. 36, pp. 1788-1795, 2003. [https://doi.org/10.1088/0022-3727/36/21/013]
• D’Alessandro, F., “Computation of dependence of lightning rod tip radius on installation height using published experi-mental results,” Proc. 28th ICLP, Kanazawa, Japan, pp. 549-554, 2006.
• D’Alessandro, F., “On the optimum rod geometry for prac-tical lightning protection systems,” Journal of Electrostatics, vol. 65, pp. 113-121, 2007. [https://doi.org/10.1016/j.elstat.2006.07.011]
• D’Alessandro, F., “Improving lightning protection with corona minimising air terminals,” Journal of Electrostatics, vol. 129, 103926, 2024. [https://doi.org/10.1016/j.elstat.2024.103926]
• D’Alessandro, F., “Characteristics and Performance of an Insulated Lightning Downconductor Cable”, Proc. 3rd In-ternat. Lightn. Protect. Symposium, Porto, Portugal, 2016.
• IEC 62305-3, “Protection against lightning – Part 3: Physi-cal damage to structures and life hazard,” Edition 2, Geneva, Switzerland, 2010.