NEDU_TR_08-01_Fish_Barrier.pdf

Navy Experimental Diving Unit LT 07-05 321 Bullfinch Rd. NEDU TR 08-01 Panama City, FL 32407-7015 June 2008

Evaluation of Risk That Electric Fish Barriers Pose to Human Immersion in the

Chicago Sanitary and Ship Canal

Navy Experimental Diving Unit

Authors:

R. P. Layton W. H. Mints V. H. Ferris

Distribution Statement F:

Further dissemination only as directed by U.S. Army Corps of Engineers, Chicago District, or higher DoD authority.

UNCLASSIFIED

SECURITY CLASSIFICATION OF THIS PAGE

REPORT DOCUMENTATION PAGE

1a. REPORT SECURITY CLASSIFICATION Unclassified

1b. RESTRICTIVE MARKINGS

2a. SECURITY CLASSIFICATION AUTHORITY

3. DISTRIBUTION/AVAILABILITY OF REPORT

DISTRIBUTION STATEMENT F: Further dissemination only as directed by U.S. Army Corps of Engineers, Chicago District, or higher DoD authority.

2b. DECLASSIFICATION/DOWNGRADING AUTHORITY

4. PERFORMING ORGANIZATION REPORT NUMBER(S)

NEDU Technical Report No. 08-01

5. MONITORING ORGANIZATION REPORT NUMBER(S)

6a. NAME OF PERFORMING

ORGANIZATION

Navy Experimental Diving Unit

6b. OFFICE SYMBOL (If Applicable)

7a. NAME OF MONITORING ORGANIZATION

6c. ADDRESS (City, State, and ZIP Code) 321 Bullfinch Road, Panama City, FL 32407-7015

7b. ADDRESS (City, State, and Zip Code)

8a. NAME OF FUNDING SPONSORING ORGANIZATION: US Army Corps of Engineers, Chicago District

8b. OFFICE SYMBOL (If Applicable)

9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER

8c. ADDRESS:

111 North Canal Street, Suite 600

Chicago, IL 60606-7206

10. SOURCE OF FUNDING NUMBERS

PROGRAM

ELEMENT NO.

PROJECT

NO.

TASK NO.

LT 07-05

WORK UNIT

ACCESSION

NO.

11. TITLE (Include Security Classification)

(U) Evaluation of Risk That Electric Fish Barriers Pose to Human Immersion in the Chicago Sanitary and Ship Canal

12. PERSONAL AUTHOR(S) R. P. Layton, W. H. Mints, V. H. Ferris

13a. TYPE OF REPORT:

Technical Report

13b. TIME COVERED:

Jan 2007 TO May 2008

14. DATE OF REPORT: May 2008 15. PAGE

COUNT 71

16. SUPPLEMENTARY NOTATION

17. COSATI CODES 18. SUBJECT TERMS: Pulsed DC electric field in water;

human shock risk; aquatic species barrier; human shock effect

FIELD GROUP SUB-GROUP

19. ABSTRACT: US Army Corps of Engineers (USACE) operates electric field—based fish barriers in the Chicago Sanitary and Ship Canal to deter invasive fish species from encroaching into the Great Lakes and Mississippi River basins.

These barriers create strong pulsed DC (4 ms pulse width, 4-5 Hz repetition rate) electric fields in the water. USACE contracted Navy Experimental Diving Unit (NEDU) to evaluate potential physiological effects to humans accidentally immersed in the water near an energized fish barrier. This report evaluates voltage gradient and contact potential data that USACE provided to NEDU to assess the level of risk that immersed humans may have to serious shock effects.

20. DISTRIBUTION/AVAILABILITY OF ABSTRACT

│ │ UNCLASSIFIED/UNLIMITED │X│ SAME AS RPT. │ │ DTIC USERS

21. ABSTRACT SECURITY CLASSIFICATION

Unclassified

22a. NAME OF RESPONSIBLE INDIVIDUAL NEDU Librarian

22b. TELEPHONE:

850-230-3100

22c. OFFICE SYMBOL

DD Form 1473 UNCLASSIFIED i

THIS PAGE BLANK

CONTENTS

Page

DD Form 1473 ........................................................................................... i Contents..................................................................................................... ii Figures ....................................................................................................... iv Tables ........................................................................................................ iv

Executive Summary ............................................................................... v

Introduction

Background Objective

Methods Approach Experimental Design and Measurements

1. Barrier I A. Voltage Gradients B. Contact Potential

2. Barrier IIA A. Voltage Gradients B. Contact Potential

Analysis Introduction Electrical Safety Standards and Codes Assessment of Shock Risks during Immersion in Canal

1. “Code of Practice for the Safe Use of Electricity Under Water, AODC 35”

A. Applicability B. Allowable Body Current

2. “IEEE Standard 80-2000” A. Applicability B. Physiological Effects of Current through the Body

3. Evaluation of Repetitive Pulse Electric Shock A. Short-Duration Repetitive Shocks B. Long-Duration Repetitive Shocks C. Effect of Water Resistivity on Severity of Shock to Immersed Body

(1) Analysis Using “Demagnetizing Factors”

Method

(2) Analysis Using Power Density Transfer

Method

4. Revised (May 2007) IEC TS 60479 Technical

Specification on Electric Shock A. Overview ii

B. Evaluation of Repetitive Short-Duration DC Pulse Electric Shock on Ventricular Fibrillation Threshold

(1) Overview

(2) Safe Body Current for Repetitive 4 ms, 5 Hz DC

Pulse Shocks during Unprotected Immersion

(3) Safe Body Current for Repetitive 4 ms, 5 Hz DC

Pulse Shocks during Garment-Protected Immersion

C. Physiological Effects of Current through an Immersed Body

5. Results from Other Studies of Electric Fields’ Effects on an Immersed Body

6. Evaluation of Risk from Contact Potential Shock Discussion

Summary of Analyses Summary of Immersion Risks in the Canal near Fish Barriers Conclusion

References Bibliography

Appendix A: Descriptions of Fish Barriers I and IIA in Chicago

Sanitary and Ship Canal ...................................................................... A1

1. Barrier I — Operational and Energized System ............................ A1

2. Barrier IIA — Operational but Nonenergized System.................... A3 iii

FIGURES

1. Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage Measured for Barriers I and IIA

2. Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage Measured for Barrier I

3. Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage Measured for Barrier IIA

4. Threshold and Limit of Ventricular Fibrillation

5. Safe Body Current Levels Plotted Versus Duration of Shock

6. Allowable AC Current in the Body

7. Thresholds for Ventricular Fibrillation from Effects of

Unidirectional Single Pulse AC Shocks of Short Duration

8. Time-Current Zones of Effects of AC Currents (15–100 Hz) on Persons for a Current Path Corresponding to Left Hand to Feet

9. Time-Current Zones of Effects of DC Currents on Persons for a Longitudinal Upward Current Path

TABLES

1A. Barriers I and IIA Maximum Electric Field Gradients and Body Currents

1B. Barriers I and IIA Maximum Contact Potentials and Body Currents

2. Time-Current Zones for Risk of Ventricular Fibrillation from Unidirectional Short-Duration, Single-Pulse AC Shocks (Summary of Zones of Figure 7)

3. Time-Current Zones for Physiological Effects of 15–100 Hz AC Single-Pulse Shock for Hand-to-Feet Pathway

(Summary of Zones of Figure 8)

4. Time-Current Zones for Physiological Effects of Single-Pulse DC Shock for Hand-to-Feet Pathway

(Summary of Zones of Figure 9) iv

EXECUTIVE SUMMARY

BACKGROUND

The U.S. Army Corps of Engineers (USACE) operates an electric field–based fish barrier (Barrier I) in the Chicago Sanitary and Ship Canal (Canal). It was constructed to prevent the migration of invasive fish species in both directions between the Great Lakes and the Mississippi River basins. A second, stronger barrier (Barrier IIA) has been constructed but is not yet operationally deployed.

The barriers consist of electrode arrays perpendicular to the water flow and lying near the bottom of the Canal. These arrays are driven by repetitive (4–5 Hz) high-voltage DC (direct current) pulses that create strong electric fields in the water. As fish approach the barrier, they receive progressively higher voltage shocks that cause them to turn away from it. In this way, the barrier blocks the passage of fish but allows boat and vessel traffic through the area.

In May 2004 an electrical discharge incident involving barge and tow vessel operations occurred near the barrier. This incident precipitated discussions between the USACE, the U.S. Coast Guard, and the U.S. Army Research and Development Center, Construction Engineering Research Laboratory (ERDC/CERL) regarding the risks posed by the fish barrier to barge and boat operations. One conclusion of these discussions was a requirement to determine the risk posed to a person accidentally immersed in the Canal near a fish barrier. To this end, ERDC/CERL made measurements on both barriers to accurately characterize the electric fields and voltages they induce in the Canal.

OBJECTIVE

The USACE contracted Navy Experimental Diving Unit (NEDU) to analyze the data obtained from the measurements performed by ERDC/CERL to determine the potential physiological effects likely to occur if a person were immersed in the Canal in the vicinity of an energized barrier. The most likely source of such an immersion would be that of an individual accidentally falling into the water from a barge, tow vessel, or recreational boat. The objective of NEDU’s evaluation of the data was to provide guidance to USACE concerning the potential severity of harmful effects to humans immersed in the Canal near a fish barrier.

APPROACH

The analysis of the effects of the measured voltages and electric field gradients and their resultant body currents on humans immersed in the Canal is complex. While many studies of the effects of electrical shocks to animals and humans are published in the v scientific literature, almost all investigate bodies in air, not immersed in water, and with single current burst shocks from 50 to 60 Hz alternating current (AC). The same is true for electrical safety standards and codes. In addition, many continuously changing environmental and physiological variables characterize the situation.

NEDU’s approach was to identify and characterize the worst-case conditions and to apply several different analytical techniques to estimate body currents. Several safety standards and codes then were extrapolated to evaluate the potential harmful physiological effects of those body currents. The goal was to bound the problem and to estimate the order of magnitude of risk for serious physiological effects from an electric shock resulting from immersion in the Canal near a barrier. Key information on the significant cumulative effect of repetitive short-duration DC shocks on the heart — information not readily available until the May 2007 revision of the IEC Technical Specification on the effect of electric current on humans and livestock — greatly increased the expected probability of ventricular fibrillation (cessation of heartbeat).

CONCLUSION

Comparing voltage gradients measured in the Canal to electric shock safety standards and to other scientific studies indicates that currents induced in a person immersed in water that includes an electric field created by a fish barrier can be life threatening.

Paralysis of body muscles, including respiratory muscles, and even ventricular fibrillation can result from such immersion. These physiological effects from electric shock could even result in death in a relatively short time. If a normal healthy adult is immersed in the electric field of a fish barrier, the most severe risk would be death from ventricular fibrillation or from asphyxia (due to either paralysis of chest wall muscles or inability to remain afloat). Once removed from the electric field, if breathing cessation is less than about three minutes and ventricular fibrillation is not triggered, individuals may require pulmonary resuscitation to survive the exposure, but normally they can recover with no long-term physiological effects. When removed from the electric field, an individual suffering fibrillation does not spontaneously recover; emergency medical treatment including defibrillation must be administered promptly (within a few minutes of initiation of fibrillation) or death will result.

NEDU’s evaluation of Canal electric field data provided by USACE indicates that the electric current densities induced in the Canal to deter undesirable aquatic species pose a significant risk to humans accidentally immersed over a Canal length of approximately 100 feet around Barrier I and 260 feet around Barrier IIA. From the reported range of water flow in the Canal, transit times for the total, but noncontiguous, 360 foot distance could keep a victim in a danger zone from 4 to 20 minutes, with ~7.5 minutes as the mean. Unless they are quickly removed from the water, persons immersed in the Canal and drifting with the current could thus spend a significant amount of time in potentially harmful electric fields. Any rescue strategy must have as its goal the removal of immersed persons as quickly as possible from the Canal, since the level of electrical body currents induced in individuals could incapacitate them within seconds.

vi

If the present input voltages to the barriers’ electrodes and the electrical current densities induced in the water already are near the minimums required to successfully prevent encroachment by invasive aquatic species, then the risk to an immersed human cannot significantly be reduced from the high level that exists from operating one or both of the barriers without compromising their effectiveness. On the other hand, if the barriers are operated at higher power levels than those currently evaluated in this report, then both an immersed person’s risk for serious physiological effects and the area of the Canal over which that risk extends will increase.

To gain perspective on the electric shock risk posed to a person immersed near a fish barrier in the Canal, such an immersion could be compared to that of an individual immersed in very cold water. In cold water immersion the victim’s survival time would likely be measured in tens of minutes; that same person immersed in the strongest electric fields of a fish barrier might survive for only minutes. Since victims in both scenarios can quickly become incapacitated and unable to aid in their own recovery, the first priority is to remove them from the water as rapidly as possible and then render appropriate emergency medical treatment. Concurrently (for the fish barriers), care must be taken to protect would-be rescuers from harm that could result if they become partially immersed in the Canal or contact immersion victims, either directly or via a metal implement, while trying to aid the victims.

The potential for accidental immersion in cold water does not prevent people from boating in such water. Similarly, the potential for serious, even fatal, harm from fish barrier electric fields will not deter everyone from entering the affected areas of the Canal. In both cases the risk exists for serious bodily injury, including death. That risk must be balanced with the overall benefit of preventing the migration of invasive aquatic species between the Great Lakes and Mississippi River basins, while at the same time allowing commercial and recreational boat traffic near the energized fish barriers in the Chicago Sanitary and Ship Canal. Appropriately informing boaters of the significant risk for serious harm from immersion in the Canal near the fish barriers, as well as implementing a robust rescue strategy, can mitigate, but cannot completely eliminate, that risk.

vii

INTRODUCTION

BACKGROUND

The U.S. Army Corps of Engineers (USACE), Chicago District, operates an electric field–based fish barrier (Barrier I) in the Chicago Sanitary and Ship Canal, Romeoville, Illinois (Canal). Smith-Root, Inc., constructed the barrier to prevent the movement of invasive fish species, such as the Asian Carp, in both directions between the Great Lakes and Mississippi River basins, and it has been operational since early 2002.1–3

Barrier I consists of six electrode arrays (each consisting of either one or four noninsulated metal conductors) perpendicular to the water flow and near the bottom of the Canal. (Detailed descriptions of the barriers can be found in Appendix A.) The electrode arrays are configured as two sets, each containing three electrode arrays.

Each set is driven by a high-voltage DC (direct current) pulse applied between the outer electrode arrays. This creates a two-lobe (Figure 1) electric field pattern in the water. As fish approach the barrier, they receive progressively higher voltage shocks, which cause them to turn away from it. At the time of Barrier I testing, the downstream set of electrode arrays were pulsed simultaneously with 800 volts (V) DC (+400 V to one array, –400 V to the second array), with 4 ms pulse width and 5 Hz repetition rate (equivalent to a 2% duty cycle). The upstream set of electrode arrays was pulsed synchronously with the downstream set but with 400 V DC (+200 V to one array, –200 V to the second array).1–3

In May 2004 an electrical discharge incident involving barge and tow vessel operations occurred near Barrier I. This incident precipitated discussions between the USACE Chicago District, the U.S. Coast Guard, and the U.S. Army Research and Development Center, Construction Engineering Research Laboratory (ERDC/CERL) regarding the risks posed by the fish barrier to barge and boat operations.1–3 One conclusion of these discussions was a requirement to determine whether this type of fish barrier is safe for barge and boat operation, a consideration including the possibility of a person accidentally being immersed in the Canal. To this end ERDC/CERL conducted electric field measurements for various barge and hull configurations passing over Barrier I.

The results and recommendations of this study have been reported.2,3

ERDC/CERL also made measurements to determine the extent of the barrier’s field and to accurately characterize the electric field and voltages to which a person immersed in the Canal near the barrier would be exposed. Details of the measurement techniques and the values of measured parameters for Barrier I have also been reported.1–3 Since the construction of the original Barrier I, a second, more powerful barrier (Barrier IIA) has been installed. While the latter is not yet approved for continuous operation, it has been energized for brief periods so that its electrical characteristics could be measured and evaluated. Some of these measured values also have been documented.3–5

Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage for Barriers I and IIA

200 0 200 400 600 800 1000 1200 1400

Distance (feet)

Figure 1. Run 17 from McInerney.4 Barrier IIA is on the left; Barrier I on the right. Voltages are measured across the probe’s electrodes, which are separated by 20 inches along each axis. Field strength, in V/ft, is equivalent to 60% of the absolute magnitude of the open circuit voltage.

Barrier I

Water Flow

Barrier IIA

Vo lta ge (v ol ts

Barrier IIA consists of four parallel arrays with a total of 32 electrodes. Two of the arrays are pulsed with 820 V DC, while the remaining two are pulsed with 498 V DC. This arrangement produces a double-lobed electric field pattern in the water (Figure 1).

Since Barrier IIA has more electrodes and more powerful pulsers than Barrier I, it can generate stronger electrical fields in the water than Barrier I can: i.e., it is capable of inducing significantly greater electric current densities in the water. The data provided for Barrier IIA were collected with 4 ms pulse width and 4 Hz repetition rate, which give a duty cycle of 1.6%. Barrier IIA pulsers can operate over a wide range of voltages, pulse widths, and duty cycles; however the parameters used when the data for this report were acquired are typical of the values planned for future operation of Barrier IIA.

A third barrier — Barrier IIB, of the same design as Barrier IIA — is planned for future construction. (Appendix A provides more information about the sizes and relative positions of the barriers.)

OBJECTIVE

The USACE contracted Navy Experimental Diving Unit (NEDU) to analyze the data obtained from the field tests of the two barriers to evaluate the potential physiological effects that are likely to occur if a person were immersed near an energized barrier in the Canal. Since no swimming or other immersed activities are permitted in the Canal, the most likely occasion of such an immersion would be that of an individual accidentally falling into the water from a barge, tow vessel, or recreational boat (i.e., a “man overboard” scenario).6

The objective of NEDU’s evaluation of the data provided by ERDC/CERL was to provide guidance to USACE concerning the potential severity of harmful effects to humans immersed in the Canal.

METHODS

APPROACH

After reviewing data from measurements conducted by both ERDC/CERL and Smith- Root, NEDU determined that they were sufficient to adequately characterize the maximum electric field strengths that an immersed person is likely to experience near either Barrier I or Barrier IIA. The data characterizing the contact potentials between the water and parts of the Canal and the barrier structures are not comprehensive, but they provide sufficient information to characterize the range of voltages and body currents to which an immersion victim might be subjected.

NEDU reviewed electric shock studies as well as appropriate electrical safety specifications and codes in the scientific literature to determine conservative, relevant physiological effects likely to occur in humans exposed to electric shock while immersed in water, with water conductivity and electric field strengths over the range found in the

Canal. NEDU also requested from Smith-Root, via USACE, copies of any other measurements, studies, or reports available to Smith-Root that would assist NEDU in evaluating the potential hazards that the fish barriers might pose to humans accidentally immersed in the Canal, but no additional information was provided.

Following conservative methods and assumptions, NEDU used the ERDC/CERL and Smith-Root measured parameters, together with generally accepted body resistance values, to evaluate the maximum electrical body currents likely to be experienced by a person immersed in the Canal. The safety and possible physiological effects of derived maximum body currents were evaluated using research studies available in the open scientific literature as well as national and international electrical safety specifications and codes.

After examining the data provided by ERDC/CERL and Smith-Root, NEDU found few resources that specifically and definitively address the situation of primary interest — the physiological effects expected in a person wearing no electrically insulative garment immersed in low conductivity (i.e., “brackish” to “fresh”) water that contains a strong electric field induced by repetitive (4–5 Hz) short duration (4 ms) DC pulses over an extended period. As a result, NEDU has evaluated the Canal’s potential human shock risks by analyzing different aspects of the problem with relatively independent techniques. In NEDU’s judgment, the best that can be accomplished by applying the available limited, hard scientific evidence in this way is to establish the order of magnitude of the risk — that is, to determine whether the maximum electric field strengths and contact voltages that ERDC/CERL and Smith-Root report in the Canal create electric currents in an immersed body that are greater than, approximately equal to, or significantly less than, the threshold currents for specific physiological effects. As will be seen, while the specific body currents examined by NEDU span a range of values for the different analysis techniques, the overall level of risk implied by all the methods is consistent.

EXPERIMENTAL DESIGN AND MEASUREMENTS

1. BARRIER I

A. Voltage Gradients For Barrier I, open circuit voltage potential measurements were made both in the horizontal plane (parallel to the water surface) and along the vertical axis. (Details of the probe and associated instrumentation have been documented.1) Mapping of the field in the vicinity of the barrier revealed that the maximum voltage gradient occurred directly over the barrier electrode arrays and was relatively uniform across the width (160 ft [49 m]) of the Canal. Since the maximum voltage potentials should be evaluated for the “man overboard” scenario in order to obtain the worst case, measurements were focused on the area above the arrays and were taken near the east bank of Barrier I.1,4,5 All the Barrier I data were collected with the barrier pulsing at ~±400 V input to the barrier’s downstream arrays, with 4 ms pulse width and 5 Hz repetition rate.1,2

In the horizontal plane the maximum open circuit voltage measured across the probe electrodes (15 inches [38 cm] apart) was 10 V, which gives a maximum calculated gradient of 8.0 volts/foot (V/ft; 26.25 volts/meter [V/m], 0.67 volts/inch [V/in], 0.26 volts/cm [V/cm]). (Data were taken from Figure 2 of Hock, McInerney, et al.1) In the vertical direction, the maximum open circuit voltage measured between the probe electrodes (6.0 ft [1.83 m] apart) was 50 V, which yields a maximum calculated gradient of 8.3 V/ft (27.2 V/m, 0.69 V/in, 0.27 V/cm). (Data were taken from Figure 4 of Hock, McInerney, et al.1)

These measurements were made during field tests in January, March, and September 2005. Following those tests, a new probe was fabricated: this consisted of three pairs of orthogonally oriented electrodes, each pair of which was 20 inches (51 cm) apart. Since with this probe the magnitude of the voltage gradient could be measured simultaneously for all three axes, it reduced or eliminated errors in total electric field magnitude, errors due to probe orientation. Figures 1 and 2 show the absolute magnitude of Barrier I (combined voltages measured for all axes) to be 17 V. The calculated maximum voltage gradient of 10.2 V/ft (33.5 V/m, 0.85 V/in, 0.34 V/cm) found with the new probe, while higher, is comparable to that found with the original probe (8.3 V/ft [27.2 V/m, 0.69 V/in,

0.27 V/cm]). The electrical parameters of Barrier I for the latter measurement were the same as those used for the original measurements: ±400 V input DC pulse, 4 ms pulse width, 5 Hz repetition rate, synchronous operation.1,2,4,5

Thus, for Barrier I the maximum field gradient measured near the surface by ERDC/CERL is 10.2 V/ft (33.5 V/m, 0.85 V/in, 0.34 V/cm). Figure 2 shows spatial aspects of this barrier’s electric field potentials.

B. Contact Potential While the above voltage gradients can be used to determine the strength of the electric field experienced by an immersed person, it also is desirable to evaluate the contact potentials that can occur when a “man overboard” touches the Canal wall or other structure, such as a rescue line or a boat, as he is climbing or being helped out of the water. The highest voltage to which an immersed person could be exposed would be the voltage between a location in the water directly over a barrier electrode array and an earth ground of the system electronics, to which the Canal perimeter fence is directly connected. This situation represents a worst-case possibility that normally would not occur unless the rescue of an immersed person was being attempted with a metal ladder, cable, or chain electrically connected directly to the Canal fence or other system ground.

The voltage differences between the perimeter fence and points along the water’s surface that are perpendicular to the Canal wall were measured for Barrier I. Data were collected from the Canal wall to a distance approximately 50 ft (15.2 m) away. The maximum open circuit voltage difference measured between a probe in the water and the Canal fence was 120 V, which could produce a whole-body pulse current of 240 mA (assuming whole-body resistance is 500 ohms [Ω]).1

Another probe was fabricated to simulate a person in the water touching the wall or other structures of the Canal. With the probe in contact with the stone wall of the Canal, a maximum current of 3.3 mA was measured. In the boat launch area, contact potential with a metal ladder produced a whole-body current of 7.4 mA when the ladder was raised out of the water but measured ~50 mA when the bottom of the ladder was submerged. Out of the water, contact with the partially submerged ladder and with a gate connected to the system ground produced currents in the range 43 to 167 mA.

Because of these high observed currents, the ladder subsequently was removed.

However, the data collected while the ladder was in position is an instructive example of how small changes in Canal structure and circumstances can have large effects on the magnitude of electric shock to which an immersed person may be exposed. While the probe was designed to mimic a person with wet hands touching an object, it did not provide for simulating a “tight” grip on any of the structure. Since a significant contribution to the electrical resistance between an immersed person and a Canal structure is this contact resistance, the measured contact potentials may be smaller than that which could be experienced by a person exerting significant effort to pull himself from the water.1

Two measurements were made of open circuit contact potential between a boat and a probe in the water. In both, the largest voltage occurred when the probe was near the boat. The maximum values reported were 35 V for the first trial and 60 V for the second.

Using the larger potential and 500 Ω whole-body resistance yields 120 mA as the DC pulse current.1

Barrier I field gradient and contact potential measurements and calculated values are summarized in Tables 1A and 1B.

Figure 2. Run 17 from McInerney.4 Barrier I voltages are measured across the probe’s electrodes, which are separated by 20 inches along each axis. Field strength, in V/ft, is equivalent to 60% of the absolute magnitude

Water Flow

Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage for Barrier I of the open circuit voltage.

2. BARRIER IIA

A. Voltage Gradients Figure 1 shows the voltages of the low and high arrays of Barrier IIA measured with the new three-axis probe. As expected, Barrier IIA’s high-voltage array creates a greater electric potential (20 V) across the electrodes of the probe than does its low voltage array (10 V). Those values were obtained with Barrier IIA pulsing the high and low voltage arrays asynchronously (820 V DC input to the array and 498 V, respectively), with 4 ms pulse width at 4 Hz repetition rate. Since these Barrier IIA measurements were taken during the same experimental run as that which produced the Barrier I data, the relative magnitudes of the two barriers’ voltage gradients should be accurate.

Thus, for Barrier IIA the maximum field gradient measured (with 20 V across electrodes spaced 20 inches [51 cm] apart) is 12.0 V/ft (39.4 V/m, 1.00 V/in, 0.39 V/cm). Figure 3 shows additional detail of the spatial characteristics of Barrier IIA’s electric field.

B. Contact Potential To determine the contact potentials, voltage differences were measured between metallic bolts screwed into the Canal wall and points on the water surface near the Canal wall. The maximum reported voltage difference was 20 V (with a measured current of ~44 mA through the probe’s 47 Ω resistance), which was recorded at some distance from the Canal wall. When the probe was adjacent to the wall, the maximum voltage measured was ~5 V with a measured current of ~6 mA.7,8 For Barrier IIA, no measurements were made of the contact potential to a metal ladder that was partially submerged nor to the perimeter fence, as had been done for Barrier I. Because of Barrier IIA’s ~18% greater maximum voltage gradients in the water, those contact potentials are expected to be greater than, but still comparable to, those reported for the first barrier.

Barrier IIA field gradient and contact potential data are summarized in Tables 1A and 1B.

Absolute Magnitude of Open Circuit, 3-Axis Probe Voltage for Barrier IIA

Figure 3. Run 17 from McInerney.4 Barrier IIA voltages are measured across the probe’s electrodes, which are separated by 20 inches along each axis. Field strength, in V/ft, is equivalent to 60% of the absolute magnitude

Water Flow of the open circuit voltage.

Table 1A. Barriers I and IIA Maximum Electric Field Gradients and Body Currents

Resistance: 470 Ω

Measured Values

Voltage Pulse Current Pulse Current

(DC Volts) (DC mA) (DC mA)

Barrier I (4 ms, 5 Hz) 10.2 V/ft 61 123 NA

Barrier IIA (4 ms, 4 Hz) High V Array 12.0 V/ft 72 144 150 Low V Array 6.0 V/ft 36 72 80

Max Measured Voltage Gradient

Large Probe (72 inches)

Limb-to-Limb* (72 inches)

Body Resistance: 500 Ω

Calculated Values

*Limb-to-limb body current was calculated from measured in-water electric field gradient as though it were applied as a contact potential (i.e., a voltage potential applied by actual physical contact between a person and an object, measured as the open circuit voltage between the two) at the end of each limb (hand or foot).

Table 1B. Barriers I and IIA Maximum Contact Potentials and Body Currents

Calculated Values

(DC Volts) (DC mA) (DC mA)

BARRIER I

Fence 120 240 Boat 60 120 Wall 3.3

Ladder Up* 7.4 Ladder Down* 49.5

Ladder Down–Gate 167

BARRIER IIA

Wall 5 6

Measured Values

Contact Potential Contact Current 500 Ω Contact Current

*The metal ladder was attached to the Canal wall. In the “up” position it did not contact the water, while in the “down” position its lower end was submerged in the Canal.

ANALYSIS

INTRODUCTION

In “EXPERIMENTAL DESIGN AND MEASUREMENTS” the ERDC/CERL measurements of barrier field strength near both Barrier I and Barrier IIA were reviewed, and the appropriate worst-case, open circuit voltage gradients were extracted. The analysis of the effects of these voltages and the resultant body currents on humans immersed in the Canal is complex. Many studies of the effects of electrical shocks to animals and humans are published in the scientific literature, but almost all investigate bodies in air, not immersed in water, and with single current burst shocks from 50 to 60 Hz AC. In addition, they address contact potential shocks (i.e., two different regions of the body contacting electrical conductors at different electrical voltage potentials), not shocks to bodies immersed in water that contains a pulsed DC electric field. As previously noted, the fish barriers produce repetitive DC pulses that create electric fields within the water to deter fish. Thus, the situation of an immersed person in the Canal is a complex one to analyze and understand, due both to the many variables that apply at any given time and to the scarcity of research studies and electrical safety standards that apply directly to the man-overboard scenario.

In the following sections, the problem is evaluated from several perspectives using different analytical approaches and techniques. The goals are to bound the problem and to estimate the order of magnitude of risk for serious physiological effects from the electric shock to which a person immersed near a Canal fish barrier may be exposed.

ELECTRICAL SAFETY STANDARDS AND CODES

Electrical safety standards, specifications, and codes are available from various regulatory bodies. Each is designed to evaluate the risk of electric shock from different segments of the electrical products industry and international communities. One fundamental point most share is that they are based on the human body’s threshold current values provided in the various editions of the International Electrotechnical Commission (IEC) Technical Specification 479. In the following sections of this evaluation, the measured values of field strength and water conductivity in the Canal will be used to estimate worst-case body currents that can be induced in a man-overboard situation. The calculated and measured current levels then will be compared to the threshold values for specific physiological effects, as elucidated in research studies and presented in versions of the IEC 479 Technical Specification for safe currents in humans.

It is important to distinguish between “threshold” currents for a shock effect and the corresponding “limit” currents. The threshold value to produce a specific physiological effect can vary widely among individuals; therefore, it is prudent to select a limit that protects most of the population. The IEC 479 specifications present current thresholds for different physiological effects. The term threshold pertains to a current level at which a certain physiological effect begins to appear in a population of individuals. A limit, on the other hand, is based on the threshold but usually includes a safety factor. As J.

Patrick Reilly explains, “In general, the more severe or life-threatening the physiologic effect, the higher the safety factor should be.”9 Figure 4 illustrates this relationship between the IEC 479 threshold currents and the Underwriters Laboratories, Inc. (UL), limit for ventricular fibrillation. The IEC document presents threshold information based on experimental work of several investigators using animal subjects. While these include the same investigations used by UL, it is important to realize that the IEC technical specification presents threshold information based on the available data and does not suggest or imply that those values be used as limits without any modification.9

In this report the threshold values for specific shock effects are used, rather than the more conservative UL limits. This is appropriate, since the UL limits are chosen to protect large populations of individuals who are regularly exposed to potential shocks from a particular source, while a man overboard in the Canal represents a severe situation to which only a few individuals may be exposed over a long time.

ASSESSMENT OF SHOCK RISKS DURING IMMERSION IN CANAL

1. “Code of Practice for the Safe Use of Electricity Under Water, AODC 035”

A. Applicability One of the communities most concerned with electrical safety in water is that involved with the use of electricity by, and around, divers. One of the most complete and authoritative documents in this community is the Code of Practice for the Safe Use of Electricity Under Water, AODC 035 (Code) published by the Association of Offshore Diving Contractors (AODC) in the United Kingdom (U.K.).10 This Code for in-water electrical safety provides the electrical safety standard sanctioned for diver use by the U.S. Naval Sea Systems Command and Naval Medical Command.11 The Code also has been adopted by the Royal Navy and is used by international commercial oil corporations concerned with risk-benefit balance and legal liability for accidents to their divers and other employees.11 This document is one of the standards used in NEDU’s evaluation of the potential risk to humans immersed in the Canal and subject to the voltage gradients measured by ERDC/CERL.

The preparation of this Code was overseen by a steering group of representatives from a cross section of interested organizations including the U.K. and other government agencies. Some of the results used to formulate the Code came from research projects funded directly by the U.K. Department of Energy. The original document was published in 1985, but a 1993 review found no amendments were necessary. Embodied in the Code is the distillation of knowledge from scientific literature, government agency engineering and research projects, and the experience of commercial and government entities regularly involved with underwater use of electricity. Since the response of living tissue to electric shock in air has been well researched, the authors of the Code incorporated that body of data and applied it to the special conditions of divers working under water.10 This extensive research makes the Code relevant to the scenario of

Threshold and Limit of Ventricular Fibrillation

C ur re nt

(m A rm s)

Shock Duration (seconds)

Figure 4. Threshold of ventricular fibrillation from IEC Publication 479-1 (1984) and limit for ventricular fibrillation from Underwriters Laboratories (1988), from Reilly.9 immersion near the Canal fish barriers.

Many years of research have been aimed at determining the levels of safety for humans in contact with electricity. Different levels of shock, causing increasing levels of physiological effects, have been identified: increasing levels of exposure result respectively in sensation, pain, involuntary contraction of muscles, and ventricular fibrillation that quickly can result in death. Yet these different levels of physiological response often overlap. And individuals exhibit different reactions to electricity: a shock that is barely perceptible to one person causes pain to another. For this reason, researchers using statistical analysis methods have relied on as great a number of subjects as is practical to determine average values. Like all biological parameters, however, susceptibility to electrical shock varies significantly among individuals, and a given level of electrical exposure exhibits a normal statistical distribution of effects.

Safety levels embodied in the Code have been set to account for the most susceptible individuals (0.5 percentile level). Factors that influence the physiological effects of a shock are variable, with the body current, duration of the shock, and weight of the individual being critical determinants. In addition, the shock response varies and depends on the current pathway in the body, the individual psychological reaction, and the frequency and mode of entry of the electric current into the body.12–38

Another widely accepted international specification on the physiological effects of electricity on the human body is Technical Specification 479 of the International Electrotechnical Commission (IEC 479), the first edition of which was published in

1974.39 The AODC 035 Code encompasses experimental work completed between the original issuance of IEC 479 in 1974 and the 1993 revision of the Code. The Code’s recommendations are based on the results of those experimental studies in combination with the physiological effects of electric current given in IEC 479.10

The level of current passing through the body subjected to electrical shock is primary in determining the effect on an individual. This is true whether the person is in air or is immersed in water. If the voltage to which the body is exposed is known, then the current can be calculated if the resistance of the body (“current route resistance”) can be determined. In the past, a value of 500 Ω has been used for a typical worst-case, limb-to-limb (right hand–to–left foot or hand-to-hand) resistance regardless of voltage.

More recent research has shown that, while this value is correct for high voltages (>50 V), it is conservative for lower voltages, where values >1000 Ω have been measured.

The Code indicates that for low voltages <50 V, 750 Ω can be safely used, while the 500 Ω value should be used for higher voltages: “Both these values are conservative and are based on the latest evidence.”10 The only exception to these values identified in the Code is the use of 100 Ω, when current may possibly be flowing from front to back of the chest due to the possibly larger surface contact path than in a limb-to-limb case.10

For a conservative evaluation of the fish barriers’ electrical effects on humans, it is reasonable to retain the traditional value of 500 Ω for whole-body (limb-to-limb, either hand-to-hand or foot-to-hand) resistance and 100 Ω for currents through the chest from front to back. Additionally, since we are considering persons in the water, the skin and contact resistances can be assumed to be minimal.10 Table 1A summarizes the maximum voltage gradients measured in proximity to the two fish barriers and the measured and calculated body currents that result.

B. Allowable Body Current Figures 5 and 6 show the curves from the Code that can be used to determine allowable body current levels for single shocks (50–60 Hz AC, or DC) of varying durations. The curves were drawn as a composite of three separate effects. If the shock is short, the only limiting value necessary is the level of current likely to cause heart fibrillation. As the shock duration increases, levels of current well below that causing fibrillation will induce involuntary muscle contraction, and in particular severe breathing difficulties, so the allowable body current curves use the lower values. When the shock duration exceeds ~10 seconds (s), an even lower value of current will produce the phenomenon in which a person cannot release an electrical conductor due to localized involuntary muscle contraction in the forearm. At this level of current, most people suffer no other short-term ill effects, since it is less than half the current necessary to cause fibrillation or breathing difficulties.10 The DC curve stops at ~200 ms, because the difference between DC and AC shocks of shorter durations than this value is of little significance. As the Code addresses risk from single current burst shock exposure of a given time duration, a significant challenge is to determine how the information from Figures 5 and 6 may be applied in the case of long-duration exposures to shocks consisting of repetitive pulses of DC voltage. This situation is important to understand, since it represents the most likely scenario for a person who may fall into the Canal.

Depending on where in the Canal such a person is immersed and on what the water velocity is, he could be exposed to the voltage gradients created by the fish barriers for many minutes.

2. IEEE Standard 80-2000

A. Applicability Another document that provides information useful in evaluating how fish barrier electrical fields may affect humans immersed in the Canal is the IEEE Guide for Safety in AC Substation Grounding, IEEE Std 80-2000.40 While this standard also focuses on the effects of a single exposure, in air, to 50–60 Hz AC, some of its guidance can be extrapolated to apply to pulsed DC electric fields in water.

Safe Body Current Levels Plotted Versus Duration of Shock

Fi br ill at in g

C ur re nt

(5 0/

H

z) E qu iv al en t (

Duration of Shock

Figure 5. Safe body current levels, from Figure 5 in AODC10.

Allowable AC Current in the Body

Fi br ill at in g

C ur re nt

(5 0/

H

z) E qu iv al en t (

Duration of Shock

Figure 6. Allowable AC Current in the Body, from Figure 13 in AODC10.

B. Physiological Effects of Current through the Body The physiological effects of an electric current passing through a given individual’s body depend on several variables: the duration, magnitude, and frequency of the current; the weight of the person; and the specific path the current takes through the body. The most dangerous consequence of such an exposure is the heart condition known as ventricular fibrillation, in which the blood immediately ceases to circulate.40

Humans are very vulnerable to the effects of electric current at commercial power frequencies of 50–60 Hz AC. Currents of approximately 100 mA root mean square (rms) for a sufficiently long duration can be lethal. (In this report all AC shocks are assumed to have sinusoid form, with the current value expressed in mA rms. All DC shocks are assumed to have rectangular waveform, with constant DC current during each pulse.)

Research indicates that in a single shock the human body can tolerate DC currents approximately five times higher than AC for a given physiological effect. The IEEE Std 80-2000, published in 2000, states that the cumulative effect of two or more closely spaced shocks has not been thoroughly evaluated, but a reasonable allowance can be made by using the sum of individual shock durations.40

In order of increasing current, the most common physiological effects of electricity on the body are threshold of perception, muscular contraction, difficulty breathing, cessation of breathing, unconsciousness, heart fibrillation, respiratory nerve blockage, and burning. The levels at which some of these effects occur are given below for 50–60 Hz AC:18–29,40

A 1 mA rms current is generally recognized as the threshold of perception, the level at which a person is just able to detect a slight tingling sensation in hands or fingertips.18

Currents of 1–6 mA rms (often called “let-go” currents), while unpleasant to sustain, generally do not prevent a person holding a charged object from being able to control his muscles and release it. For the 0.5 percentile population, 6 mA rms for women and 9 mA rms for men are the measured let-go threshold values.23

Currents of 9–25 mA rms may be painful and may make it difficult or impossible for the hand to release energized objects it has grasped. For still higher currents, muscular contractions could make breathing difficult. The effects of 9–25 mA rms currents usually are not permanent and disappear when these currents are removed, unless contraction is very severe and breathing is stopped for minutes rather than seconds.19

Currents of 60–100 mA rms can cause ventricular fibrillation, heart stoppage, or cessation of respiration — and result in permanent injury or death.20,21

Thus, if a victim is to survive, the magnitude and duration of the maximum allowable current conducted through a human body at 50–60 Hz at least must be less than the value that can precipitate ventricular fibrillation.40 Experiments on animals have shown that two factors primarily govern the intensity of current needed to induce ventricular fibrillation: body weight and shock duration. For single-pulse shocks (50–60 Hz AC) of a duration between 0.03 and 3.0 s, it is accepted that 99.5% of all persons can safely withstand, without fibrillation, the passage of a current with magnitude and duration given by

IB = k x (tB s) , –1/2 where IB is the current in amperes rms, ts is the duration in seconds, and k is a constant dependent on subject weight. For 99.5% of persons weighing 50 kg (110 lb), the equation becomes

IB = 0.116 x (tB s) . –1/2

This result has been derived for humans from the empirical data collected from animal studies. For 1 s shocks, the corresponding current is 116 mA rms; for 0.1 s shocks, the corresponding current is 367 mA rms.13,17,40 However, the IEEE Standard 80-2000 states that the relation may be more complex, as some recent experiments suggest the existence of two distinct thresholds: one in which the shock duration is shorter…