SPECIAL ISSUE: ENVIRONMENTAL SOCIAL IMPACT ASSESSMENT (ESIA) AND RISK ASSESSMENT OF CRUDE OIL AND GAS PIPELINES

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 1-8 Published online October 15, 2015 (http://www.sciencepublishinggroup.com/j/ijema) doi: 10.11648/j.ijema.s.2015030601.11 ISSN: 2328-7659 (Print); ISSN: 2328-7667 (Online)

Quantitative Risk Analysis for Gas (NG and NGL) Pipelines Huseyin Murat Cekirge Department of Mechanical Engineering, the Grove School of Engineering, the City College of the City University of New York, New York, USA

Email address: [email protected]

To cite this article: Huseyin Murat Cekirge. Quantitative Risk Analysis for Gas (NG and NGL) Pipelines. International Journal of Environmental Monitoring and Analysis. Special Issue: Environmental Social Impact Assessment (ESIA) and Risk Assessment of Crude Oil and Gas Pipelines. Vol. 3, No. 6-1, 2015, pp. 1-8.doi: 10.11648/j.ijema.s.2015030601.11

Abstract: By using index method and multivariable analysis, a methodology of the threat and risk of natural gas (NG) and natural gas liquids (NGL) pipelines to environment will be presented by considering total infrastructure. General concepts are introduced and explained in detail. Keywords: Quantitative Risk Analysis, Risk Valorization, Risk Analysis of NG Pipelines, Risk Analysis of NGL Pipelines, Acceptable Risk Zones, Unacceptable Risk Zones

1. Introduction A gas pipeline presents risk to individuals and environment in general. The frequencies of occurrence of threatening events and distances at which the highest probability of risk is reached are the components of the risk analysis. These components proceed to calculate the maximum individual risk for each scenario considered. Subsequently, this level risk can be compared with other daily activities that involve risk. The maximum individual risk is the increased chance of death of a person exposed to a threat in a period of time. The calculation of the maximum individual risk at (x, y) considers all threatening events that can be generated by: • Jet Fire (Jet-Fire): It occurs when there is continuous leakage of highly pressurized flammable gas is turned to jet fire near the point of the leak. A stream of fire generally produces thermal radiation continuous. The size of the affected area depends on the discharge rate of the gas, orientation and direction of the jet, and prevailing weather conditions at the time of the event. • Vapor Cloud Fire (Flash Fire or Flare): The flare corresponds to the rapid ignition of a vapor cloud. It occurs in the area between the lower and upper limits of flammability of vapor cloud, where appropriate air-vapor ratio in contact with flammable source ignition causes combustion of the available mass, generating thermal radiation. It is worth noting that the mass contained in the cloud is not enough to generate overpressure.

• Vapor Cloud Explosion (Unconfined Vapor Cloud Explosion): It happens after the release of a large amount of flammable gas or vapor to atmosphere with ignition point some distance from the exhaust, causing a sudden release and violent energy in waves of pressure. A necessary condition for cloud explosion or gas pressure wave generation is caused by the presence of containment or obstruction, which favors high burning rates. • Geotechnical stability: In the process of operation of gas pipeline, it has implemented a maintenance program to ROW (Right of Way) that is the result of assessment of the sensitivity the pipeline. It has been done an initial assessment, with which they have taken the first steps to reduce the risk from the point of view of geotechnical stability of the ROW, valuation constantly updated to ensure that maintenance activities are executed in terms of reducing the risk of gas pipeline due to geotechnical stability The main issue in this paper is the methodology and data needed to prepare risk analysis of total infrastructure of a gas pipeline. General principles of the methodology, the specific equations, and data required to prepare a risk analysis are discussed, and an example is presented to illustrate the method. The references considered in this paper are issued by the Federal Emergency Management Agency (FEMA) for

2

Huseyin Murat Cekirge. Quantitative Risk Analysis for Gas (NG and NGL) Pipelines

emergency planning, the U.S. Environmental Protection Agency (EPA) for environmental concerns, the U.S. Department of Transportation (DOT), the U.S. Department of Commerce (DOC), and various publications by the Center for Chemical Process Safety (CCPS) of the American Institute of Chemical Engineers, [1- 14]. These principles have been modified specifically for pipelines in the context of this paper. A fundamental aim of this paper is to present an assessment method that is providing reasonable risk estimates for policy decisions and certain assumptions used are part of the paper. The methodologies presented in the paper cover pipelines carrying natural gas (NG) and natural gas liquids (NGL). The data required for a risk analysis includes pipeline data and site data. Some information that would aid in a risk analysis is proprietary to the pipeline operator. In general, the required data include: The location of the proposed pipeline site, including roads and major terrain feature boundaries; The location of the pipeline with respect to the proposed site, and specifically the segment lying within the site; Land use and terrain characteristics adjacent to and within the site; The pipeline diameter, operating pressure, and for liquid pipelines, the product flow rate; and Pipeline operating history information, especially records of any previous accidental releases of product and the repair history, if available. A phase of environmental assessment study will identify the near site and several key characteristics of a hazardous pipeline such as: • Location; • Product transported; • Diameter; • Operating pressure; • Materials of construction; and • Date of construction. The paper provides for two types of calculations involving the individual risk: • Estimating the individual risk at a specific distance from the hazard source of a pipeline segment; and • Estimating the distance corresponding to a specified level of individual risk. The fundamental approach in the paper is the former, as described in detail in the remainder of this section. The paper also describes the latter, which can be done through the basic process by iterating on distance as described briefly later in this section. The steps of a risk analysis, in sequence, determine the: 1. Hazard impact distance; 2. Segment length for hazards based on the distance between the receptor and the pipeline hazard source, and the hazard impact distance; 3. Maximum mortality impact from the closest approach of the pipeline to the receptor; 4. Average mortality at the receptor for each segment, 5. Base adjusted failure probability for the pipeline; 6. Base probability for each segment;

7. Conditional probability factor for each event scenario, 8. Conditional probability of individual exposure; and 9. Individual risk at the specified locations. Appropriate hazard consequence modeling of product releases is the basis for estimating the hazard impact distances. The scenarios apply for each of the hazard categories previously stated, i.e., flash fires, jet fires (pool fires for liquid releases), and unconfined gas or vapor explosions. The data needed for the uses of this part of the evaluation include the following: • Product transported by the pipeline; • Pipeline diameter; • Pipeline operating pressure; • Minimum distance between the pipeline and the property line (or boundary between the unusable portion and usable portion of a site, which may apply to some sites); • Orientation of the pipeline to the property line (i.e., parallel, perpendicular, at an angle, etc.); • Length of property line exposed to pipe length of concern, the length of the pipeline segment that lies within 500 meter of the property lines; and • The receptor location distance as the center of the property line nearest to the pipeline.

2. Features of Risk Assessment System This document contains the results of the risk analysis of technological pipeline transport system, for which a methodology has been applied to meet all specifications that are required by the risk management of the infrastructure. It has been estimated as a basis for risk valorization and as a tool for the determination of the most dangerous areas in the event of a leak or spill which has not been declared a threatening event type of fire, but for the existence of gas vapors and liquid product and predicting the effects demand to take appropriate actions. Through the methodology used, it is also to have a basis for risk management, represented by a risk value for each section of the pipeline, and defined from the threat related variables, the constructive aspects of the system and the events that may originate from third parties or natural phenomenon. The study explains step by step the aspects of the necessary methodology that the operating group responsible for maintaining the “Contingency Plan” information makes necessary modifications to update the risk analysis, and also use these results in decisions for annual investment programs, conditioning it to risk management in the system. This study initially present general aspect of risk, to facilitate understanding of the procedures and results of the methodologies employed in selection and estimation causes and risk. Such methodologies are described later and are carried thereon to application pipeline transport system. The valuation methodology is based on the universal concept of risk, in which the parameters are determined. The risk is amended through semi quantitative leveraging estimation consequences to quantitative values and qualitative assessments that define some parameters or aspects of risk.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 1-8

3. Risk Assessment Methodology The methodology used is a adaptation of indexing method developed in [15] and [16] using influence distances reported for the estimation of consequences and characteristics of operation pipeline transport system. The risk variables, balance from the point of view of individual risk covered by Guidelines for Chemical Process Quantitative Risk Analysis that includes advances in Chemical Process Quantitative Risk Analysis (CPQRA), [17 - 25], are modified for using them in the analyses properly. The methodology incorporates risk valorization specific data input and output that must be updated constantly by staff in each infrastructure. It is the input characteristics that modify the risk, which are related to infrastructure, operation, maintenance, environment, prevention activities, and the characteristics of the load. These characteristics influence the frequency of a novice event, the probability of occurrence of a threatening event, the probability of producing damage in the area in which the event occurred given in the same conditions and environmental vulnerability. These characteristics will be valued for the conditions of the pipeline transport system by sector, so that each will produce a separate segmentation in the corridor. The combination of the values of these characteristics affected by factors defined for each weight results as a result a risk factor for the sector, which will be represented along the pipeline to facilitate analysis. The methodology allows the calculation of a risk value of all infrastructures per unit length, besides being able to view the areas of greatest risk along the pipeline. The risk is defined as the combination of four factors, one of which is a frequency (occurrences per unit time) and others are likely to occurrence (dimensionless values) of events that are attached in a threatening event, resulting in a risk value in terms of the occurrence of a particular damage per unit time. To facilitate risk management defines a set of characteristics that infer in the same and related infrastructure, ancillary systems, the operation, safety programs and environmental

3

characteristics. These characteristics determine the risk, the four factors are: • the frequency of occurrence of an event (f Ei); • the probability of a threatening event arising since the initiating event (PrEa); • the probability probability of occurrence an effect with potential damage given the environment (PrEf); and • the probability that a given event effect that threatening damage occurs (Pr Damage). With these values defined for each sector, probability of occurrence of damage given the vulnerability of the area. The characteristics analyzed represent damage considered in the risk assessment.

4. Characteristics of Risk Factors of a Gas Pipeline The performing of valorization can be done in the following: • identify partitions that each feature occurs along the pipe and ; and • assign the respective values of each sector using sectorization maxima occur in the system. Each feature referred to faith will be given a relative value, which is a number between 0 and 10 following the guidelines presented in Table 1 and considering each risk characteristic independently. Threatening events considered are fire, jet fire and explosion blaze. The unconsidered event, that is not included among these events, it is said that the event is a consequent pollution of NG and NGL scattering or dispersion of the spill. The features should be evaluated in the area of influence of the infrastructure, which is defined by the distance of influence to the situation of the pipe. The risk factors are listed in the following table:

Table 1. Basic risk factors for fEi . Variable

Definition

C1

Age of the pipe

C2

Installation

C3

Protection of corrosion

C4

System security for pipe pressure

C5

Dangerous activities in the area

Value VC 20 years and more = 20 Less than 20 years = years/2 Open = 10 (unprotected) Open = 4 (protected) Buried less than one meter = 8 (unprotected) Buried less than one meter = 4 (protected) Buried more one meter = 4 (unprotected) Buried more one meter = 0 (protected) Note: The protection concerns structural coatings prevent the action of efforts. No corrosion = 10 One system of corrosion = 7 Two system of corrosion = 4 Three system of corrosion = 0 Two systems to prevent corrosion = 4 Three or more systems to prevent corrosion = 0 Note: No one considers the lining of the pipe as a system to avoid or prevent corrosion. The minimum value is zero and the maximum is = 10. Localization_1 = 8; Localization_2 = 5; Localization_3 = 3 and Localization_4= 1. Intensive= 10 Moderate = 8

4

Huseyin Murat Cekirge. Quantitative Risk Analysis for Gas (NG and NGL) Pipelines

Variable

Definition

C6

Social conflicts in the area

C7

Natural disaster in the area

C8

Maintenance

C9

Security inspections

C10

Social programs

Value VC None = 0 Note: Refers to construction activities, planting, excavation or earthworks. Terrorism = 10 Moderate social conflicts = 4 Minimal conflicts = 0 Level_1 = 10 Level_2 = 8 Level_3 = 4 Level_4 = 0 4 * Number of natural threats It must consider the existence of threats by movements earth phenomena hydro meteorological, erosion, processes basins, seismic and volcanic processes. The minimum value is zero and the maxim or 10. No activity = 10 Moderate activity = 8 High level maintenance = 0 Note: activities should be considered as suspended pumping a sector block, operation Valve, restart operations, maintenance of the pipeline, keeping valves, and auxiliary systems maintenance. The minimum value is zero and the maximum 10. No activity = 10 Moderate activity = 8 High level inspection program = 0 No activity = 10 Moderate activity = 8 Some activity = 4 Continuous communication = 0 Note: Each year, the program implemented should include at least one divulgence for the community capacitating in the entire area of influence.

In Table 2, the basis for the valuation of the characteristics referred to the likelihood of a threatening event since the beginner event; probability of an effect is present for potential damage is (PrEa* PrEf). The threatening events considered are jet fire and explosion flare. If none of these events is presented, the consequent event is pollution and dispersion of NG; or spill and dispersion NGL will be actual. The characteristics must be assessed in the catchment area of infrastructure, which is defined by the distance from involvement of pipeline conditions. Table 3 presents guidelines to assess the material characteristics of the

probability of harm (PrDamage). With these values defined for each sector, probability of occurrence of damage is the vulnerability of the area. Table 2. Features regarding PrEa and PrEf . Variable

Definition

C11

Points that may cause ignition

Value VC VC11 = 10 Cpi /(0.11 e (7.67 (Dpi / Daf)))(1) Cpi = number of infrastructures Dpi = distance of the infrastructures Daf = distance of the influence The minimum value is zero and values greater than 10 will be 10

Table 3. Features for calculating PrDamage . Variable

Definition

C12

Exposed people

C13

Areas of environmental importance

C14 For NGL pipes

Areas of emergency

C15

Training programs

C10

Social programs

Value VC VC12 =10 Cp /(0.80 e (4.56 (Dpe / Daf))) (2) Cpe = number of people Dpe = distance of the people The minimum value is zero and values greater than 10 will be 10 VC13 =10 Caa /(0.11 e (7.67 e (7.67 (Daa / Daf))) (3) Caa = number of special areas Daa = distance of the these areas The minimum value is zero and values greater than 10 will be 10 VC14 = 10 Cap /(0.11 e (7.67 (Dap / Daf))) (4) Cap = number of potential areas Dap = distance of the potential areas The minimum value is zero and values greater than 10 will be 10 No training activity = 10 Moderate training activity = 8 High level training program = 0 Note: A yearly program implemented and address: a strategic level divulgation two disclosures and training at the tactical level and three disclosures and trainings at the operational level. No activity = 10 Compliance with some activities = 8

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 1-8

Variable

Definition

C16

Reporting

C17

Alarm system

5

Value VC Program implemented = 0 Note: Each year, the program implemented should address by least divulgence training to the entire community area of influence. Systematic reporting system = 0 Moderate reporting = 4 No reporting = 10 Well established alarm system = 0 Moderate alarm system = 6 No alarm system = 10

5. Risk Calculation by Sectors The value assigned to each feature is used to calculate the respective value of frequency or probability of each risk factor, using the formulas presented in Table 4. Risk information by sector is plotted along the length abscissa of the pipe in order

to easily identify the areas of greatest risk index. The incidence values of each feature as shown in Table 5, which builds on the statistics of incidents in the transport of hydrocarbons, the U.S. Department of Transportation (DOT), [6 - 8], with adjustments for the similar activities that developed in the world.

Table 4. Calculation of risk factors. Variable

Frequency of Beginning Event

Probability of an Event Threatening Environmental Conditions Probability of Damage of Threatening Event

Risk for Sector

Calculation fEi = 0.1 fi Σ Vc Ic (5) fEi = factor for of beginning of an event fi = initial frequency(= 9.68 10-6 year -1 for pipes with diameter 2 and less inches; = 5.56 10-6 year -1 for pipes with diameter between 2 and 6 inches and less inches and = 2.688 10-6 year -1 for pipes with diameter more 6 inches) Vc = characteristic value Ic = indice value of the characteristic, Table 5. Pr Ea= 0.026 VC12

(6)

PrHuman Damage =(1- Dp / Daf) 2.67 Cp (1 + VC14/10) (1- VC15+VC10+VC16+VC17)/160) (7) PrArea Damage =(1- Dp / (0.5476 Daf) ) 2.67 Ca (1 + VC14/10) (1- VC15+VC10+VC16+VC17)/160) (8) RsHuman Damage = fEi PrEa PrHuman Damage (9) RsArea Damage = fEi PrEa PrArea Damage (10) For each sector of the risk assessment for each type of damage and these values are represented independently for each sector.

Because the input ranges are not the same, the pipeline transport system must be analyzed for the number of partitions that go to these ranges of combined characteristics of each risk factor. The value of influence distance must not be less than half of ROW (Right of Way) established for the pipeline. The influence distances are ROW for scenarios that result in lower influence distance values. Table 5. Incidents referred to FEi. Characteristic Age of the pipe Pipe Installation Systems to avoid or prevent corrosion Factor System Security Threat activity in the area Threat social Threat natural phenomena Procedures applied for and maintenance operation Procedures implemented safety inspections Programs socialization

Indices 0.039 0.039 0.189 0.039 0.270 0.270 0.039 0.038 0.038 0.039

6. Maximum Individual Risk Individual risk involves the likelihood, that particular stage

in a threatening event at a point x, y, is assumed for calculations that the individual is at the point x,y; for 24 hours a day, 360 days of the year, which would be the most critical situation. In numerical terms; the individual risk is defined as the probability of death of a person or a number of them, for a specific area in a year. The calculation is as shown as: IR

x,y,i=

f

EA,i

P

f,i

(11)

where; IR x,y,i= individual risk at x, y of threat of i, f EA,i=frequency of threat i in a year and P f,i = probability of effect of threat i to a person at x, y. The Pf,i is determined from the percentage of assignment given by probability and for practical purposes is equivalent to dividing by total occurring events, that is the percentage. Two terms are used in the calculation of individual risk, only the probability Pf,i is a function of geographic position. Therefore, the maximum individual risk is reached at point x, y; and that influence results a probability of 1.00. Mathematically, the maximum individual risk is equal to the frequency value of threatening event f EA,i. The total maximum individual risk at x, y; is given by the sum of the individual maximum risks that may arise in different scenarios and their respective threatening events:

6

Huseyin Murat Cekirge. Quantitative Risk Analysis for Gas (NG and NGL) Pipelines

potential areas, and their respective distances of involvement are (Dpi, Dpe, Daa and Dap). The characteristics C11 to C14 are calculated through these values. The results of the other characteristics of the study assessed the risk of the product are presented in Tables 8 and 9. By considering Tables 10 and 11, intermediate frequency of the event data likelihood of harm to people and environmental can be seen through the probability of initiating events, which are fire, blaze, and dispersion. Based on the values on these tables, the values of the total risk of affecting people and environment from NG and NGL pipelines are assessed for risk acceptability for each sector of the pipelines, Tables 12 and 13.

(12)

where; IR x,y =total risk of n threatening effects at x, y and IR x,y,j =total risk of jth threatening effects at x, y. Tables 6 and 7 present the results of evaluation of risk values of pipeline transport system applying the criteria described in the paper. The values (Npi, Npe, Naa and Nap) are modifying the risk of ignition, exposed, environmental and

Table 6. Characteristics referred to initiating an event of NG pipeline. No

Sector

C1

C2

C3

C4

C5

C6

C7

C8

C9

C10

C15

C16

C17

1

KP 0+000 to 0+004

1

4

4

8

0

0

0

0

0

4

8

4

10

2

KP 1+888 to 1+898

1

2

4

8

8

0

0

0

0

4

8

4

10

3

KP 4+264 to 4+274

1

2

4

8

8

0

0

0

0

4

8

4

10

4

KP 6+056 to 6+764

1

4

4

8

8

0

0

0

0

4

8

4

10

5

KP 8+700 to 8+710

1

2

4

8

8

0

0

0

0

4

8

4

10

Table 7. Characteristics referred to initiating an event of NGL pipeline. No

Sector

C1

C2

C3

C4

C5

C6

C7

C8

C9

C10

C15

C16

C17

1

KP 0+000 to 0+004

1

4

4

8

0

0

4

0

0

4

8

4

10

2

KP 1+964 to 2+341

1

4

4

8

0

0

4

0

0

4

8

4

10

3

KP 4+358 to 4+491

1

4

4

8

0

0

4

0

0

4

8

4

10

4

KP 6+825 to 6+835

1

2

4

8

0

0

4

0

0

4

8

4

10

5

KP 8+788 to 8+888

1

4

4

8

0

0

4

0

0

4

8

4

10

Table 8. Features regarding NG Pipeline Vulnerability. No 1 2 3 4 5

Sector KP 0+000 to 0+004 KP 1+888 to 1+898 KP 4+264 to 4+274 KP 6+056 to 6+764 KP 8+700 to 8+710

Npi 2 1 17 17 17

Dpi 200 500 225 225 225

Npe 8 1 67 67 67

Dpe 200 500 225 225 225

Naa 1 1 1 1 1

Daa 100 10 10 50 10

Nap 2 1 1 1 1

Dap 50 50 50 50 50

C11 0 0 1 1 1

C12 2 0 8 8 8

C13 3 10 10 10 10

Table 9. Features regarding NGL Pipeline Vulnerability, C14 are included. No 1 2 3 4 5

Sector KP 0+000 to 0+004 KP 1+964 to 2+341 KP 4+358 to 4+491 KP 6+825 to 6+835 KP 8+788 to 8+888

Npi 8 1 17 17 17

Dpi 200 500 225 225 225

Npe 8 1 67 67 67

Dpe 200 500 225 225 225

Naa 1 1 1 1 1

Daa 50 50 50 10 50

Nap 2 1 1 1 1

Dap 50 50 50 50 50

C11 6 0 0 10 0

C12 6 0 1 10 0

C13 10 6 8 10 4

Table 10. Results of NG pipeline risk assessment, blaze values are 0.263 of fire values. No

Sector

1 2 3 4 5

KP 0+000 to 0+004 KP 1+888 to 1+898 KP 4+264 to 4+274 KP 6+056 to 6+764 KP 8+700 to 8+710

Fire Risk Person 5.56E-11 3.59E-14 2.65E-12 2.70E-12 2.65E-12

Risk Environment 6.46E-10 6.94E-09 7.64E-09 4.47E-09 7.64E-09

Blaze Risk Person 1.47E-11 9.50E-15 7.00E-13 7.16E-13 7.00E-13

Risk Environment 1.71E-10 1.84E-09 2.02E-09 1.18E-09 2.02E-09

C14 10 6 8 10 4

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 1-8

7

Table 11. Results of NGL pipeline risk assessment, blaze values are 0.263 of fire values, C14 is considered for dispersion. No

Sector

1 2 3 4 5

KP 0+000 to 0+004 KP 1+964 to 2+341 KP 4+358 to 4+491 KP 6+825 to 6+835 KP 8+788 to 8+888

Fire Risk Person 3.53E-09 1.29E-14 9.74E-13 1.71 E-07 7.57E-13

Risk Environment 1.21 E-09 2.89E-10 3.72E-10 2.16E-09 2.03E-10

Blaze Risk Person 9.33E-10 3.42E-15 2.58E-13 4.54E-08 2.00E-13

Risk Environment 3.19E-10 7.66E-11 9.86E-11 5.71 E-10 5.37E-11

Dispersion Risk Person 1.17E-08 7.86E-14 5.93E-12 4.14E-07 4.61 E-12

Risk Environment 4.02E-09 1.76E-09 2.27E-09 5.21 E-09 1.24E-09

Table 12. NG pipeline risk acceptability, all columns are summed. No

Sector

Risk People

Risk Environment

1 2 3 4 5

KP 0+000 to 0+004 KP 1+888 to 1+898 KP 4+264 to 4+274 KP 6+056 to 6+764 KP 8+700 to 8+710

7.0E-11 4.5E-14 3.3E-12 3.4E-12 3.3E-12

8.2E-10 8.8E-09 9.7E-09 5.7E-09 9.7E-09

Acceptability of Risk Risk Person Acceptable Acceptable Acceptable Acceptable Acceptable

Risk Environment Acceptable Acceptable Acceptable Acceptable Acceptable

Table 13. NGL pipeline risk acceptability, all columns are summed. No

Sector

Risk People

Risk Environment

1 2 3 4 5

KP 0+000 to 0+004 KP 1+964 to 2+341 KP 4+358 to 4+491 KP 6+825 to 6+835 KP 8+788 to 8+888

1.6E-08 9.5E-14 7.2E-12 6.3E-07 5.6E-12

5.5E-09 2.1E-09 2.7E-09 7.9E-09 1.5E-09

If the total individual risk is calculated less than the risk in traffic accidents, the pipeline can be considered as safe infrastructure and in acceptable risk zone. It should be mentioned that risk of cancer death which is 1.96 * 10-3 per year, and less than the risk of death traffic accidents is 3.9* 10-4 per year, Table 14. Table 14. Individual risk of early fatality by various causes, [26]. Hazard Heart Attack Cancer All Accidents Motor Vehicles Homicide Drowning Fire Civil Aviation Release of gas from Wahsatch pipeline/well network (risk level - 100 meters either side of pipeline or wellhead) Water Transport Railroad Accidents Lightning Bites And Stings Release of gas from Wahsatch pipeline/well network (risk level - 100 meters either side of pipeline or wellhead) Release of gas from Wahsatch pipeline/well network (risk level - 300 meters either side of pipeline or wellhead)

Approximate Individual Risk of Death One Chance Probability / Year in Years 3.12*10-3 320 1.96*10-3 510 3.90*10-4 2560 1.98*10-4 5030 8.66*10-5 11500 2.09*10-5 47600 1.93*10-5 51600 5.19*10-6 192600 5.20*10-6

192,300

3.90*10-6 2.56*10-6 4.06*10-7 2.79*10-7

256000 389000 2454000 3570000

2.00*10 -7

5,000,000

2.50*10 -9

400,000,000

Acceptability of Risk Risk Person Acceptable Acceptable Acceptable Acceptable Acceptable

Risk Environment Acceptable Acceptable Acceptable Tolerable Acceptable

7. Conclusions The introduction of qualitative risk of gas pipelines can be seen in [27], where the risk of gas pipelines was explained in detail. In this present study, the estimation of risk of a pipeline transport system is presented for assessing the individual risk and total risk based on the characteristics of the gas pipeline and its operations. These evaluations are based on results of estimation of threatening consequences of risk events identified. Individual risk assessment has its foundations in the application of procedures used in other transport infrastructure of hydrocarbon industry. In these procedures, all spill events that have occurred to date are included. In considering risk valorization characteristics of a pipeline transport system and its operations, the process valorization has been defined for individual and environmental risk. If the risk values of each infrastructure are located in acceptable risk levels, then the infrastructure could be considered acceptable. The sectors have been identified with acceptable risk for the human settlements are safe sectors of the infrastructure. In the area of human settlements, there should be divulgation activities and preparation of communities for preventive measures. The ground motions in the view of integrity of the infrastructure should be monitored by bi-annually and annually or in shorter periods through scheduled activities. The necessary updates must be introduced and included in old and new maintenance programs.

8

Huseyin Murat Cekirge. Quantitative Risk Analysis for Gas (NG and NGL) Pipelines

References [1]

FEMA, Emergency Response Guidebook, 2008.

[2]

FEMA, DOT, and EPA, Handbook of Chemical Hazards Analysis Procedures, US Environmental Protection Agency, US Department of Transportation, and the Federal Emergency Management Agency, 1989.

[3]

United States Environmental Protection Agency (EPA), Evaluation of Dense Gas Simulation Models, Draft, EPA 450/4-90, United States Environmental Protection Agency, September 1990.

[4]

United States Environmental Protection Agency (EPA), Risk Management Program Guidance for Offsite Consequence Analysis (OCAG), EPA 550-B-99-009, United States Environmental Protection Agency, April 1999.

[5]

United States Environmental Protection Agency (EPA), TANKS 4.09b, Computer Program, US Environmental Protection Agency, September 1999.

[6]

United States Department of Transportation Office of Pipeline Safety (DOT-OPS), Gas Pipeline Incident Database, 1984-200 1, and Hazardous Liquid Incident Database, 1986-2001.

[7]

United States Department of Transportation Office of Pipeline Safety (DOT-OPS), 2000 Annual Reports for Natural Gas Transmission and Distribution Pipeline Operators.

[8]

United States Department of Transportation, Research and Special Programs Administration, 49 CFR Part 192, [Docket No. RSPA00-7666; Amendment 192-95] RIN 2137-AD54, Pipeline Safety: Pipeline Integrity Management in High Consequence Areas (Gas Transmission Pipelines), ACTION: Final rule.

[9]

U.S. Department of Commerce (US DOC), "Heat radiation from Large Pool Fires" NISTIR 6546, Fire Safety Engineering Division Building and Fire Research Laboratory, November 2000. Risk Assessment, American Institute of Chemical Engineers, New York, 1995.

[10] Frank Lees, Hazard Identification, Assessment and Control, Loss Prevention in the Process Industries Second Edition, 1996. [11] Eltgroth, Mark W., "CHARM® Emergency Response System Technical Reference Manual," Radian Corporation, September 1995. [12] Radian International, CHARM® Technical Manual, 1995. [13] HSE, "The Effect of Explosions in the Process Industries,"

Loss Prevention Bulletin, 68:37-47, Health & Safety Executive (HSE), 1986. [14] HSE, Second Report Advisory Committee Ma/or Hazards, U.K. Health and Safety Commission, Health & Safety Executive (HSE), 1979. [15] W. K, Muhlbauer, Pipeline Risk Management Manual, Second Edition, Gulf Publishing Co., Houston, TX, 1996. [16] W. K, Muhlbauer, Pipeline Risk Management Manual, Third Edition, Gulf Publishing Co., Houston, TX, 2004. [17] Risk Assessment, American Institute of Chemical Engineers, New York, 1995. [18] CCPS (Center for Chemical Process Safety),Guidelines for Fire Protection in Chemical, Petrochemical, and Hydrocarbon Processing Facilities, 2003. [19] CCPS (Center for Chemical Process Safety), Guidelines for Chemical Process Quantitative Risk Analysis, 1999. [20] CCPS (Center for Chemical Process Safety), Guidelines for Evaluating the Characteristics of Vapor Cloud Explosions, Flash Fires, and BLEVES, American Institute of Chemical Engineers, New York, 1994. [21] CCPS (Center for Chemical Process Safety), Guidelines for Evaluating Process Plant Buildings for External Explosions and Fires, American Institute of Chemical Engineers, New York, New York, 1996. [22] CCPS (Center for Chemical Process Safety), Guidelines for Chemical Process Quantitative Risk Analysis, American Institute of Chemical Engineers, New York, New York, 1989. [23] CCPS (Center for Chemical Process Safety), Guidelines for Hazard Evaluation Procedures, American Institute of Chemical Engineers, New York, 1992. [24] Greenwood, B., L. Seeley, and J. Spouge, "Risk Criteria for Use in Quantitative Risk Analysis," in CCPS, International Conference and Workshop on Risk Analysis and Process Safety, October 2124, 1997, Atlanta, Georgia, American Institute of Chemical Engineers, New York, New York, 1997, pp. 29-40. [25] McAllister, E. W. (ed), Pipeline Rules of Thumb Handbook, Gulf Publishing Co., Houston, Texas, 1993, pp. 497-501. [26] John B. Cornwell and William E. Martinsen, Quantitative Risk Analysis of the Wahsatch Gas Gathering Pipeline System, Houston, Texas, 1994. [27] Huseyin Murat Cekirge. Qualitative Risk of Gas Pipelines. American Journal of Energy Engineering. Vol. 3, No.3, 2015, pp. 53-56. doi: 10.11648/j.ajee.20150303.14

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 9-17 Published online October 15, 2015 (http://www.sciencepublishinggroup.com/j/ijema) doi: 10.11648/j.ijema.s.2015030601.12 ISSN: 2328-7659 (Print); ISSN: 2328-7667 (Online)

Determination of Safety Zones of Gas (NG and NGL) Pipelines Huseyin Murat Cekirge Department of Mechanical Engineering, the Grove School of Engineering, the City College of the City University of New York, New York, USA

Email address: [email protected]

To cite this article: Huseyin Murat Cekirge. Determination of Safety Zones of Gas (NG and NGL) Pipelines. International Journal of Environmental Monitoring and Analysis. Special Issue: Environmental Social Impact Assessment (ESIA) and Risk Assessment of Crude Oil and Gas Pipelines. Vol. 3, No. 6-1, 2015, pp. 9-17. doi: 10.11648/j.ijema.s.2015030601.12

Abstract: Quantitative analyses are necessary to find a safe passage of natural gas (NG) and natural gas liquids (NGL) pipelines. The paper is presenting the details of analyses and determination of safe zones around the gas pipelines. The methodology is a planning tool for determining the safe passages for gas pipelines. The methodologies are presented with examples. The necessary tools are also presented for these analyses. Keywords: Safety Zones of Gas Pipelines, Safety Zones of Gas Pipelines, Safety Zones of NGL Pipelines

1. Introduction The gas pipelines have various threats to environment [1] and [2], and for safety location of these pipelines, a quantitative analysis is a must to determine the safe passage of these pipelines through urban or other sensitive areas. The paper will present these analyses of fire and explosion for optimum design and proper location of the gas pipelines. The understanding of any event threatening the area of by gas pipeline is necessary. The treat is the release of hydrocarbon from the pipeline, and this event event is called accident. The event of accident is defined as release or loss of material and/or energy contained in the pipeline. According to this definition, the event identified of this study is the leaking of hydrocarbons from NG and NGL pipeline. From the time when the accident occurs, it will be developed one or more threatening events, [3-11]. The type and amount of threatening events depend on the hydrocarbon in the pipeline: • The characteristics of mass flow and pressure of the product at the site and time of breakage; • Discharge conditions; • The influence of the receiver on the generation of hazardous event; and • The atmospheric conditions. Additionally, in the development of threatening events NGL pipeline, involving the following characteristics: The medium (soil or water) receptor spill; and Spread features of spills in this medium.

Discharge conditions depend greatly on the size of the opening break. For this study, three types of releases are considered: • Hole, • Rupture; and • Total Rupture which considers discharges through holes with diameters equivalent of 5%, 25% and 90% of the total diameter of the pipeline, respectively. The accidents are: Pool Fire: It happens if volatile vapors spilled product fractions within the upper and lower limits of flammability, contact with an ignition source, transferring sufficient energy to generate a fire of the whole mass of product is located a threatening event of interest. The resulting thermal radiation can generate a domino effect on vulnerable or chained elements adjacent area. The duration of the fire is related to the nature and quantity of fuel available to burn. The hydrocarbon fire may occur after the spill; in this case the spill can be confined by a dike or topography. Jet Fire: It occurs when there is continuous leakage of highly pressurized flammable gas is turned near the point of the leak. A stream of fire generally produces thermal radiation continuous. The size of the affected area depends on the discharge rate of the gas, orientation and direction of the jet, and prevailing weather conditions at the time of the event. Vapor Cloud Fire (Flash Fire or Flare): The flare

10

Huseyin Murat Cekirge: Determination of Safety Zones of Gas (NG and NGL) Pipelines

corresponds to the rapid ignition of a vapor cloud. It occurs in the area between the lower and upper limits of flammability of vapor cloud, where appropriate air-vapor ratio in contact with flammable source ignition causes combustion of the available mass and generating thermal radiation. It is worth noting that the mass contained in the cloud is not enough to generate overpressure. Unconfined Vapor Cloud Explosion (UVCE): It happens after the release of a large amount of flammable gas or vapor to atmosphere with ignition point at some distance from the exhaust, causing a sudden release and violent energy in waves of pressure. A necessary condition for cloud explosion gas pressure wave is caused by the presence of containment or obstruction, which favors high burning rates. Geotechnical Stability: In the process of operation of the pipeline, it has implemented a maintenance program of the ROW (Right of Way) that reflects an assessment of the sensitivity. It has been done an initial assessment, with which they have taken the first steps to reduce the risk from the view of stability of the ROW, evaluations constantly updated to ensure that maintenance activities are executed in terms of reducing the risk of the pipeline. Infrastructure breakage may occur due to different factors or causes. It defined an assessment of integrity of the infrastructure from the standpoint of geotechnical stability, which has taken into account the constructive process, environmental conditions, and the phenomena of the natural environment and the level of risk. It represents the evaluated parameters of infrastructure to proximity population centers, difficult access to the site and nearby rivers with population. To set the distance to which spilled material is extended must be calculated with consideration of evaporation of NGL in the cases NGL pipelines. Considering expansion to restricting water bodies, it is assumed that the product could stay twice as long at a distance of 70 kilometers, regardless of the product can stay on the banks of the stream water. In product storage conditions to form a pool, the product could have a longer residence time, in which case it would not be traveling on a spill route, and it may form 1000 barrels of a pool of 900 square meters. The evaporation rate would be approximately 4x10-5 meters per hour, which means that the product would have greater permanence in a pool form if the

product does not infiltrate.

2. Threatening Events, Accidents 2.1. Determination of the Volume Flow from NG and NGL Pipelines The amount of the gas discharge from a NG pipeline is limited by the pipeline valves according to location. Given the types of breakage, valves before and after the cleavage site would isolate the sector. The volume is governed by the volume of gas content and pressure when a rupture happens, except for the first two minutes until the valves are being shut off. In minor breakages, this type of discharge is remained powered by the system’s working pressure. The initial gas discharge flow will be defined mainly by initial system pressure at the time of breakage. This flow will decrease as gas is being dislodged as the internal pressure of gas equals to atmospheric pressure, at which discharge ceases. This applies to discharges due to a full and half break. Table 1 provides a summary of the results of valuation discharge from NG pipeline. The final results of the analysis of consequences depend directly on the speed at which the product discharges to atmosphere. This speed is a function of operating pressure at the time of breakage in the case of the NG pipeline, and additionally for the NGL pipeline for transport, the speed of discharge depends on the maximum height difference of the product column. Initially, the release rate increases due to the pressure difference existing between the operating pressure at the point of rupture and pressure atmospheric. In the case of the spill or leak, the pressure difference in the discharge decreases to zero. At this moment the discharge starts to be governed only by the rate of pumping. For the case the NGL, by the static height of the column of liquid that may drain the site breakage. Discharge product tends to stabilize over the flow of normal operation, if the automatic closing valves do not act. When valves close in the NGL pipeline, the unloading is the product corresponding to the volume of content between the valves and the breaking point. Table 2 presents a summary assessment NGL discharge.

Table 1. The flow from the NG pipeline. No. 1 2 3 4 5

Location Part 0 10 20 30 40

(km+m) 0+000 10+071 19+854 30+030 39+999

Level meter 378.9 423.9 552.4 705.3 631.4

Diameter inches 32 32 32 32 32

Pressure (psi) 1812.5 1807.3 1802.2 1797.0 1791.8

Flow Rate Min. (kg/s) 12.4 12.3 26.3 26.2 26.2

Med. (kg/s) 61.8 61.7 131.5 131.2 130.8

Max. (kg/s) 222.7 222.0 473.5 472.2 470.8

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 9-17

11

Table 2. The flow from the ruptured NGL pipeline. Level

DIAMETER

VOLUME

SPILLED MATERIAL

Part

(km+m)

meter

NOMINAL inches

Spill barrels

Min. Daily

Med. Hourly

Max. Minute

1

0

0+001

382

14

717

2.5

2.4

11.9

2

10

10+164

405

14

383

1.4

1.3

6.5

3

20

19+932

548

14

2437

5.9

5.7

26.6

4

30

30+084

752

14

479

1.9

1.9

9

5

40

40+047

611

14

1409

3.2

3

14

No.

Location

Table 3. Levels of Protection and involvement by thermal radiation, [12]. THERMAL RADIATION (KW/m2) > 37.5 > 20.9 > 14.50 9.50 - 12.50 7.25 – 9.50 5 – 7.25

1.6 – 5

6.4 > 3.25 >3 2-3 0.4 – 2 < 0.4

DESCRIPTION Maximum peak overpressure cannot develop a bang confined hydrocarbon vapours. This level of overpressure does not cause death, but if it reaches a probability of impact 45% from ruptured eardrum. > 6.4 almost complete destruction of houses. Possible damage of tanks storage and processing equipment. Chance of involvement 10% from ruptured eardrum. The eardrum rupture threshold (1% probability) is presented this overpressure. Inside this area occur in severe damages on steel and masonry structures (industrial buildings). Within this zone, the partial collapse of roofs occurs and walls of houses. Pressure levels sufficient to cause minor damage to structures, houses and buildings. Area exposed to levels below 0.4 psig pressure; 50 percent domestic broken glass. The probability, that there is no severe damage in limit, is 95 percent and sets the distance of security for the population before the explosion.

2.4. Vapor Cloud Explosion NG and NGL The effects of an overpressure of an explosion which reaches the person can be fatal. If the person is away from the edge of the cloud bursts, the pressure is unable to cause death directly, but indirectly. This is the case of a pressure wave that can collapse a structure, which falls on a person. The death of the person is a result of a collapse of the structure would not be directly inevitable if the is in an open area. In the event of an explosion of a vapor cloud, the harm to the public is

determined function overpressure levels, regardless of the exposure time. As people exposed to a peak overpressure do not have time to react or protect themselves. One effect of a pressure peak on people is ruptured eardrums, so the distances in which there are 50, 10 and 1 percent probability is evaluated chance of this type of damage. [16–18]. Table 11 presents ranges overpressure characteristic values. In Figure 2. we can see the representation of these values.

Figure 2. Impact areas by explosions, [2].

3. Quantifying Threat 3.1. Spill Frequency Identification The threat of a leak or spill flammable product described in terms of the frequency of occurrence of threatening events

affect the environment. In general, performing a quantitative risk analysis for transport operations of dangerous substances is conditional on an acceptable analysis of frequency of incident, for which historical data required for frequency calculations. For the pipelines,

14

Huseyin Murat Cekirge: Determination of Safety Zones of Gas (NG and NGL) Pipelines

Base Frequency =Number of failure / (Length of the Pipeline *Year)

(1)

is called base frequency; as this is obtained from international data, and therefore does not consider the situations of public order and social life that may affect the operation of the pipelines. Obtaining records frequency of leaks in transmission lines can be obtained from statistics reported by international control organizations. The wide range of international statistics can be summed up employing Equation 2, which relates the calculation of the base frequency or base escape rate with the factors listed below: Base Frequency * (Number of failure / (Length of the Pipeline*Year)) = fage fproduct fdiameter fzone

(2)

Variables in Equation 2 are determined from the relationships presented in Tables 12 and 13, [19] and [20]. Table 12. Age factor pipe. SIZE OF LEAK Small Medium Large Total

fage 7.66 + (age – 6.5) * 1.18 1.16 + (age – 6.5) * 0.179 0.776 + (age – 6.5) * 0.119 0.0621 + (age – 6.5) * 0.00954

Table 13. Factors determining base frequency. Material Crude

fproduct 1

fdiameter

White Products

0.46

( 103.5 –2.9 * ϕ ) / 71

NG/NGL

0.35

Zone Open Area

fzone

Urban

1

0.32

ϕ = Diameter, inches

Table 14 provides the base frequency to the operating conditions and layout the pipeline. All previous calculations considered the value of 0.35 for fproducto, assigned to transport of NG and NGL or a similar product, and considering the greater part of the path in rural areas, through a pipe 14" and 32" (crude oil pipeline and gas pipeline, respectively) with an operating age of 2 years. The values reported above are totally statistics from other systems within the past two years of operation. They have been presented in the 5 events in the pipeline operation, which determines spills of 220 km of NGL pipeline in two years; it gives a frequency of 1.14 * 10-2 spills / Km Year. In order to make an adjustment real and actual frequency are averaged with the statistics in the final risk assessment. The frequency versus escape breaks can be set, and the distribution is presented in Table 15.

Table 14. Obtaining the base frequency. Rupture

Age

Substance

SMALL MEDIAN MAXIMUM Total

2.35 0.3545 0.2405 0.01917

0.35 0.35 0.35 0.35

Diameter NG Pipeline 0.15 0.15 0.15 0.15

SMALL MEDIAN MAXIMUM Total

2.35 0.3545 0.2405 0.01917

0.35 0.35 0.35 0.35

NGL pipeline 0.89 0.89 0.89 0.89

Table 15. Base frequency distribution according to the size of rupture. Size of the Rupture Leak Hole Rupture

Percentage of occurrence 89.0 10.0 1.0

3.2. Estimation of Probabilities Event The frequency of occurrence of a threatening event is determined mainly by the probability of existence of ignition sources that create such events and operating conditions of the discharge. The ignition probability depends on three factors namely: Mass involved in the event; Temperature of the substance released; and Existence of ignition source.

Area

Event / (km year)

0.32 0.32 0.32 0.32 Total frequency

2.46E-05 3.72E-06 2.52E-06 2.01E-07 3.11E-07

0.32 0.32 0.32 0.32 Total Frequency

1.45E-04 2.19E-05 1.48E-05 1.18E-06 1.83*10-4

The event tree technique is a graphical logic model that identifies and quantifies potential hazardous threatening events developed after the spill or leak of flammable substance. The event tree provides coverage of systematic sequence in time of the development of the threat, as are the weather conditions, presence of ignition sources, etc. Within the procedure for estimating risk, the utility of event tree is determining threat that is in the intermediate probabilities of the event initiation, Figure 3. Based in Figure 3, the frequencies of occurrence of the events are obtained for threatening events. The analysis required to obtain the probability of occurrence of each threatening event which are for pool fire, flare and cloud explosion.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 9-17

15

Figure 3. Threating tree analysis of a rupture.

Therefore, the frequency of occurrence of events, fire and flame, f FIRE and f FL, respectively, will be given as f f

= f accident P

fire fl

(3)

i

= f accident (1 - P i ) P

(4)

it

where; P fl = (1 - P i ) P it , Probability of flame; P i = Probability of ignition of the discharge phase; P it = Probability of delayed ignition; and P e = Probability of explosion. 3.3. Calculating the Probability of Ignition Ignition corresponds to source of energy that can make an escape gas or vapor generated by a flammable liquid materialize to a fire. The ignition can occur both at the point of escape or any distance from this point. Estimated values of the probability of ignition, whether as ignition or specific ignition sources, estimated for leaks flammable fluids (gases or liquids), can be seen in Table 16, [21] and [22]. Table 16. Estimation of the probability of ignition. Escape from Rupture SMALL (< 1 Kg/s) MEDIUM (1 – 50 Kg/s) MAXIMUM (> 50 Kg/s)

GAS

LIQUID

0.01 0.07 0.30

0.01 0.03 0.08

4. Assessment of Individual Risk 4.1. Risk Estimation Once, the frequencies of occurrence of threatening events are obtained, then distances at which the highest probability of death can be calculated. The maximum individual risk for each scenario considered can be calculated. Subsequently, this level risk can be compared with other daily activities that involve risk. The maximum individual risk is the increased

chance of death of a person exposed to a threat, within a period of time. The calculation of the maximum individual risk at (x, y) considers all threatening events that can be generated by fires and flares. For the calculation of the maximum individual risk, the following probabilities are taken into account for simplifications: The probability of wind direction in the sense that it can affect a person is 0.5. This considering is that the wind is in all directions. The probability of death by contacting jet flame is the maximum (100%), the lower limit of this distance is the distance of flammability of flare of the event; and it is the probability of death contact direct to the jet fire; radiation is above 37.5 kW / m2. As each threatening event generates a maximum distance; the final distance taken into account for the highest individual risk is lower than risk of all other events that is considering the three types of breaks, fire, flame, and explosion. The calculation of the maximum individual risk is given by; R

xyij

= fExposition fDi PIji LSECTIONji PWind_direction

(5) (6)

i = types of failure considered. j = type of fire (pool fire and flare) fDj = (number of spills or leaks / year- Km); base frequency by type of spill; x, y = location of break j; PIji = probability according to type of threat (j), according to the type of failure (i). LSECTIONji = length in kilometers of pipelines considered for the study, and it is section is determined based on the maximum and both sides or double sides of the corridor of hazard generated any of the events (j) point x, y.

16

Huseyin Murat Cekirge: Determination of Safety Zones of Gas (NG and NGL) Pipelines

Table 17. Individual Risk Levels, by considering Equation 5, fExposition=1. FREQUENCY

PERCENT LEAK

Pr EVENT, PIji

DISTANCE, LSECTIONji

Probability of Wind Direction, PWind_direction

Individual Risk, R xyij

3.11E-05 3.11E-05 3.11E-05 3.11E-05 3.11E-05 3.11E-05

0.89 0.89 0.1 0.1 0.01 0.01

0.07 0.0651 0.07 0.0651 0.07 0.0651

62.5 45 230.6 229 314 1632

0.125 0.125 0.125 0.125 0.125 0.125

1.51E-08 1.01E-08 6.28E-09 5.80E-09 8.54E-10 4.13E-09 4.23E-08

3.11E-05 3.11E-05 3.11E-05 3.11E-05 3.11E-05 3.11E-05

0.89 0.89 0.1 0.1 0.01 0.01

0.07 0.0651 0.07 0.0651 0.07 0.0651

61.4 43.6 240.9 235 365 2058

0.125 0.125 0.125 0.125 0.125 0.125

1.49E-08 9.82E-09 6.56E-09 5.95E-09 9.93E-10 5.21E-09 4.34E-08

STATISTICS, fDi Km 08 + 800

TOTAL Km 50+600

TOTAL

PWind_direction = probability of wind direction to the person located at the point x, y and fExposition = factor determines the probability that the person at the point (x, y), who is exposed to the event of danger averaged according day and night. The person is exposed to the event of hazard; the situation is averaged day and night. For calculating the exposure factor; fExposition = (PED (1- POUT) PREDUCTION + PEN PFIRE (1- POUT) PREDUCTION) / 2 (7) where; fExposition = The probability that the person is at x, y; PED = Probability that the person is on site daily, is considered as 0.69; PEN = Probability that the person is in the room at night, is

considered as 1.0; POUT = Probability that the person is outside the house, is considered as 0.35; and PREDUCTION = Reduction factor generated by being the person inside the house and as a conservative factor, 0.45 The above values determine an exposure factor as 0.54. Table 17 presents the maximum individual risk levels calculated for the points more frequent events. 4.2. Analysis of Results According to the information presented in Table 17, the presence of the discharge generates an individual risk with values between 4.23* 10-8 and 4.34 * 10-8 deaths / year to the areas adjacent to the corridor line.

Figure 4. Individual risk levels.

Since there is no official regulation to evaluate and determine levels tolerable and acceptable risk, in this study a comparison is established with national and international statistics on usual pre-existing risks, as well as levels of acceptability adopted by different countries. By considering general HSE (Health, Safety and Environment) rules, the schematic in Figure 4 sets out the risk levels identified where

more than 10-3 value is an intolerable risk to working activities considered "dangerous"; for a person outside the industry limit tolerability of risk decreases to 10-4. Values between 10-3 and 10-6 are considered tolerable; only if necessary steps are taken to reduce risks to levels that are reasonably practical. Values less than 10-6 are considered negligible, and the risk classified as acceptable. Based on the criteria of Figure 4 and the risk

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 9-17

17

levels reported in Table 17 identify that all points in this section of the pipeline, have levels of tolerable risk.

[10] GIUSP, The Gas Industry Unsafe Situations Procedures–Edition 6, Including 1 & 2, 2 April 2012.

5. Conclusions

[11] Young-Do Jo and Daniel A. Crowl, Individual risk analysis of high-pressure natural gas pipelines, Journal of Loss Prevention in the Process Industries, Volume 21, Issue 6, Pages 589–595, November 2008.

The methodology is presented is detailed analysis of determinations of safety zones of NG and NGL pipelines for accidents. The analysis is covering fire, flame and explosions. In addition to these analyses, estimation individual risk and safety are discussed. This report is guidance to design the ROW for a gas pipeline; however a system must be established to determine safety zones during accidents by considering prevailing topographic, environmental and meteorological conditions. This is necessary to minimize the fatalities.

[12] Guidelines for Chemical Process Quantitative Risk Analysis. AICHE, Second Edition, 2000. [13] Steve Lewis, An Overview of Leading Software Tools for QRA, American Society of Safety Engineers –Middle East Chapter (161), 7th Professional Development Conference & Exhibition, Kingdom of Bahrain, www.asse-mec.org, March 18-22, 2005. [14] OGB, Risk Assessment Data Directory, International Association of Oil & Gas Producers, Report No. 434 – 7, March 2010.

References

[15] http://www.risk-support.co.uk/consequence_modelling.ht m, 2015.

[1]

Huseyin Murat Cekirge. Qualitative Risk of Gas Pipelines. American Journal of Energy Engineering. Vol. 3, No.3, 2015, pp. 53-56. doi: 10.11648/j.ajee.20150303.14

[2]

H. M. Cekirge, Safety Analysis, Maltepe Uni., Internal Rep. 1/1, Istanbul, 2010.

[16] R. Michael Reynolds, ALOHA (Areal Locations of Hazardous Atmospheres) 5.0 Theoretical Description, NOAA Technical Memorandum NOS ORCA-65, Seattle, Washington 98115, August 1992.

[3]

Guidelines for Chemical Process Quantitative Risk Analysis. AICHE. Second Edition.2000.

[4]

FEMA, Emergency Response Guidebook, 2008.

[5]

FEMA, DOT, and EPA, Handbook of Chemical Hazards Analysis Procedures, US Environmental Protection Agency, US Department of Transportation, and the Federal Emergency Management Agency, 1989.

[6]

United States Environmental Protection Agency (EPA), Evaluation of Dense Gas Simulation Models, Draft, EPA 450/4-90, United States Environmental Protection Agency, September 1990.

[7]

J. L. Kennedy, Oil and Gas Pipeline Fundamentals, Pinacle Books, Richford, VT, 1984.

[8]

Rizkalla, M. (editor), Pipeline Geo-Environmental Design and Geohazard Management, ASME, 2008.

[9]

W. K. Muhlbauer, Pipeline Risk Management Manual: Ideas, Techniques, and Resources, Gulf Publishing Services, Burlington, Mass., 2004

[17] United States Environmental Protection Agency (EPA), ALOHA®ArealLocations of Hazardous Atmospheres), User's Manual, United States Environmental Protection Agency, Office of Emergency Management, Washington, D.C. and National Oceanic and Atmospheric Administration, Office of Response, Seattle, Washington, February 2006 [18] United States Environmental Protection Agency (EPA), Download of ALOHA, Version 5.4, http:www.epa.Rov/ceppo/cameo/request.htm. Accessed September 2006. [19] J. R. Taylor, Review of Failure Rate Data for Risk Analysis, 1998. [20] J. R. Taylor, Risk Analysis for Process Plant, Pipelines and Transport, E & FN Spon, New York, 1994 Guidelines for Chemical Process Quantitative Risk Analysis. CCPS, AIChE, 1989. [21] A. W. Cox, Frank P. Lees and M. L. Ang, Classification of Hazardous Locations, IchemE, 1990. [22] Frank Lees, Loss Prevention in the Process Industries. Hazard Identification, Assessment and Control. 1996.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 18-25 Published online October 15, 2015 (http://www.sciencepublishinggroup.com/j/ijema) doi: 10.11648/j.ijema.s.2015030601.13 ISSN: 2328-7659 (Print); ISSN: 2328-7667 (Online)

Response Strategy and Scenarios for Accidents in Crude Oil and Gas Pipelines Huseyin Murat Cekirge Department of Mechanical Engineering, the Grove School of Engineering, the City College of the City University of New York, New York, USA

Email address: [email protected]

To cite this article: Huseyin Murat Cekirge. Response Strategy and Scenarios for Accidents in Crude Oil and Gas Pipelines. International Journal of Environmental Monitoring and Analysis. Special Issue: Environmental Social Impact Assessment (ESIA) and Risk Assessment of Crude Oil and Gas Pipelines. Vol. 3, No. 6-1, 2015, pp. 18-25. doi: 10.11648/j.ijema.s.2015030601.13

Abstract: This paper introduces response scenarios used for spill training, planning, and real time oil spill response to be utilized for mitigation of spill accidents. A series of scenarios with the response guidelines and strategies are presented during accidents of crude oil and natural gas (NG) and natural gas liquids (NGL) pipelines. Keywords: Accident Scenarios for Hydrocarbon Pipelines, Risk Scenarios, Strategies for Accidents of Crude Oil Pipelines, Strategies for Accidents of NG and NGL Pipelines

1. Introduction The response scenarios for accidents happening hydrocarbon systems are important part of planning, training and real time mitigation plans of hydrocarbon pipeline systems. Each system has own properties and its special situation during the accidents. In the study, sample scenarios and its necessary elements will be introduced, and that will a starting point for real-like and appropriate accident scenarios. As it is consistent across the entire Pipeline System, the Operating Company as an Operator has adopted the internationally recognized three-tiered approach for classification of oil spills in the design of its oil spill response capability. For each tier, a planning volume has been established which corresponds to a volume that can reasonably be expected for the pipeline in the area. Tier descriptions, and their corresponding planning volumes are summarized as follows for off-shore and on-shore spills. Tier 1: Minor Spills Tier 1 includes small operational spills which can be responded to by local personnel. Based on the most probable spill size, the Pipeline General Oil Spill Plans may set a planning volume of approximately 500 m3 for pipelines and approximately 300 m3 for marine spills. Tier 2: Major Spills Tier 2 includes incidents that can be responded to with in-country resources. The planning guideline volume for Tier 2 pipeline spills is are based on the Quantified Risk

Assessment Report which identifies environmental risk associated with different pipeline failure modes and with the highest environmental risk (based on possible spill volume, and potential frequency) associated with pipeline holes (50 mm) caused by third party accidents. Based on these numbers, the Tier 2 spill planning guideline volume of approximately 3000 m3 has been adopted for pipeline spills in the pipeline area. For marine spills, the Pipeline General Oil Spill Plans may set the Tier 2 value at 2,000 m3; 1 m3 is considered to be approximately equal to 1 t. It is important to recognize that in actual practice, the Tier 2 volume will be determined by incident-specific parameters and may differ from this planning volume. Tier 3: Crisis Event Tier 3 requires augmenting in-country resources with assets from other countries. Although extremely unlikely, a Tier 3 incident will most likely result from a full bore rupture of the pipeline or a tanker collision/grounding. The Pipeline General Oil Spill Plans may set the Tier 3 planning value around 100,000 m3 for a marine related incident. The Tier 3 planning volume may not set for land pipeline spills in the Pipeline Company’s General Oil Spill Plan, although the average calculated full-bore rupture loss is approximately 6 000 m3. A rapid response is essential in the containment and control of an oil spill. The Operating Company has structured its response capability to meet Response Time Planning Guidelines, which cover from spill notification, through mobilization of initial resources and their arrival at

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 18-25

deployment locations, and the time to deploy Tier 2 Full Capability resources, if needed. The Operating Company has trained emergency response personnel contactable by telephone, pager and/or radio on standby 24 hours a day, 365 days a year. Response time planning guidelines are summarized in Table 1 and are consistent across the entire Pipeline System. For each response depot, a mobilization planning time of 2 hours is anticipated for initial responders to report to an assigned

19

response base. Response equipment located at a response base will be stored and maintained to allow for a planned departure time from a response base within 2 hours of arrival of personnel. Response depots are located within 4 hours driving time of any control point under daylight and clear weather conditions. Delays due to snow or other adverse conditions are unpredictable and cannot be provided for in these response time planning guidelines.

Table 1. Pipeline response time guidelines. Activity Notification Mobilization at Response Depot Departure from Response Depot with appropriate equipment Travel time to spill site Deployment of initial response resources at spill site Tier 2 Full Capability in place using equipment and resources from other response bases

Considering the size and complexity of the pipeline area as well as logistical factors, response depots must be established to meet the response planning guidelines. The location and number of the response depots are depending on response time and pipeline characteristics. It should be noted that airport with access to these depots is very important in the case Tier 3 spills. The airports should facilitate landing of large plane that will carry heavy response equipment. The location of equipment depot requires detailed analysis for optimum response, the equipment and and personnel stationed in these depots must be determined according to a detailed analysis of whole pipeline, surrounding environment and containment sites, [1 - 4]. 1.1. Response Strategy On-Shore Spills The oil spill response strategy considers a Tiered response, beginning with the ability to handle the spill with on-site personnel (Tier 1), progressing to a Tier 2 spill utilizing resources internal to the Operating Company and the area, and extending to Tier 3 when additional (external) resources are needed. The strategy to respond to land-based spills utilizes the following components that are in place along the pipeline: Establishment of a Tier 1 capability and appropriate spill response kits at each facility including Personal Safety Equipments (PPE); Placement of response kits on each truck coming to a facility; Employment of a Response Contractor to provide 24-hour Tier 2 response capability for the pipeline and terminal system; Participation in an international Tier 3 response organization; Identification of minimally two downstream Containment Sites for the capture and recovery of potential oil spills from each one-kilometer segment of pipeline; Establishment of a number of response depots designed to meet planning transit time guidelines discussed

Response Time (hr) 0 2 2 4 4 12

Cumulative (hr) 0 2 4 8 12 24

previously; Selection of equipment to meet the following time, spill volume recovery, and storage guidelines, 1 m3 = ~1 t : Less than 12 hours for the deployment of First Response resources sufficient to remove 600 m3 in 36 hours at each primary Containment Site; Less than 24 hours for the deployment of a Full Tier 2 recovery capability of 3000 m3 to be in place within 24 hours; and A First Response less than 12 hours; in-place storage capacity of 360 m3 of which 25% will be handled by portable storage tanks and 33% by lined pits, and considering 40% losses between the pipeline and Containment Site plus a 100% emulsification factor, and the remaining material will be removed by tank truck or other means. 1.2. Response to Marine Spills For the Marine Terminal, key elements of the response strategy include the response considerations listed above with the following additions: The strategy is focused on response to spills at the terminal and its surrounding waters; Spill containment and recovery at the loading terminal, including encirclement of part or the entire vessel by retention boom as necessary; Marine Oil Spill tracking system, [5, 7] On water containment and recovery, utilizing a specially designed Oil Spill Response Vessel (OSRV), and tugboats outfitted with a boom and skimmer system; Open water containment; using open water boom to contain drifting oil to be collected by an on water skimming system; On water storage, using special barges to store recovered oil; Placement of equipment at designated Containment Sites to prevent the oiling of interior marine habitats; Shoreline protection, including the use of nearshore and shoreseal boom, filter fences and weir systems to prevent

20

Huseyin Murat Cekirge:

Response Strategy and Scenarios for Accidents in Crude Oil and Gas Pipelines

the contamination of sensitive interior waters and shoreline; Nearshore and shoreline recovery; to collect spilled oil using a variety of skimmers designed for different uses and oil types; and Shoreline cleanup; including the equipment and analysis of preferred methods to remove oil from all shoreline types present in the area. The following time and spill volume recovery guidelines are used: Less than 12 hours for the deployment of First Response resources must be sufficient to remove 300 m3 in 48 hours at a single Containment Site; Less than 24-hours for the deployment of a Full Tier 2 recovery capability of 2200 m3 must be in place from two response depots within 24 hours. The Operating Company will implement actions to cleanup oil released from its facilities to reasonable levels. No standards for determination of how-clean-is-clean for oil spills are known to exist in the area, or internationally. The Operating Company will work with the government, and others as appropriate, to establish guidelines and/or realistic target cleanup levels prior to emergency need. 1.3. Response to NG and NGL Pipeline Accidents The Operating Company will establish equipment and response depots along the NG and NGL pipelines that are different than crude oil pipeline. Fire, flame, explosion, spilling and toxication [8 - 19] are the components of the accidents. The selection of locations of these response depots must be appropriate to the limits of the safety guidelines. These depots must be equipped fire and explosion safety equipment, modeling capabilities to establish fire and explosion safety zones [20 -22]. The rescue teams for fire and explosion must be established and the depots must be connected national and international centers that are in this nature. The response equipment list must be prepared by considering the specifications and terrain of the pipeline.

2. Response Scenarios 2.1. Response Scenarios of Crude Oil Pipelines Response scenarios illustrate the manner in which the spill response may be undertaken, and these scenarios will be developed for use in training and drills exercises. The issues are: Activation Guidance; Containment Site location by KP (Kilometer Point); Containment Sites ; and Containment Site Reports and EM (Environmental Map) maps; Equipments Available; Potential Spill Quantity for Pipeline and Marine; Spill Assessment and Tracking; Response Depot and Time to Respond: Rivers and Streams at the Spill Source;

River and Stream Velocities; Sensitivity of Downstream Receptors; Special Sensitive Areas at Spill Source; Sources of Information Used in Scenario Development Equipment Database; and Notification Database and other related databases. The following sample scenarios are selected: Scenario 1 - Lubrication Oil Spill at Workshop in Pump Station PUMP-1, September; Scenario 2 - Valve Leak at Pigging Station at Pump Station, July; Scenario 3 - Pipeline Rupture KP 290, March; and Scenario 4 - Large Release from Vessel Cargo Tanks, August. Scenario 1 - Lubrication Oil Spill at Workshop in Pump Station PUMP-1, September. Situation Report: Lubrication oil drum falls from forklift while being moved; Location: KP 21.86; Drum is righted but half spills inside the workshop; Habitat type(s): Concrete floor and no oil leaves the workshop; Probable Spill Size: 60 liters; Month: September and Oil Characteristics: Medium viscosity lubes oil. Response Level: Tier 1 Response Objectives: Protect worker health and safety; Stop the leak and repair the drum; Stop the spread of oil; Remove oil; and Properly dispose of wastes. Response Activity (by Hour) 0-1 hours: Leak controlled. Sorbent boom and pads deployed from Tier 1 spill kit. 1-2 hours: Cleanup continues with sorbents and rags. 2-3 hours: Final oil removed by pressure wash. Oily waste collected. 3-4 hours: Cleanup terminates. Additional activities: The Operating Company employs contractors for waste removal to certified landfills. Equipment: Tier 1 spill kits (local) and Workshop employees use sorbents and PPE from the Tier 1 Spill Kits. Scenario 2 - Valve Leak at Pigging Station at Pump Station PUMP-3, July Situation Report: Valve flange leak at pigging station at Pump Station PUMP-3; at KP 444.98 and

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 18-25

Immediately brought under control, but oil sprayed over surrounding area. Habitat type(s): Dirt, gravel, and paved road. No oil leaves the facility. Probable Spill Size: 30 m3. Month: July. Oil Characteristics: Flowing, medium viscosity. Response Level: Initial Tier 1 – fully uses all material in Tier 1 Spill Kits and Tier 2 – Activated. Response Objectives: Protect worker health and safety; Stop the leak; Prevent spread of oil into the environment; Complete government and downstream industry notifications; Remove pooled oil; Remove contaminated soils and gravels; Remove oil from hard surfaces and Properly dispose of wastes. Post Cleanup: Restore the area with clean material. Response Activity (by hour) 0-1 hours: Leak controlled, Tier 1 cleanup initiated and Response Contractor notified and prepares accordingly. 1-2 hours: Response Contractor called out, departs equipment depots: Tier 1 equipment exhausted. 2-3 hours: Cleanup continues with shovels. 4-6 hours: Response Contractor arrives (travel time 3 hours), deploys equipment, Cleanup continues, use of sorbents, shovel cleanup, and bagging of oily material and Hard surfaces cleaned with high-pressure washers. 7-8 hours: Cleanup terminates. Additional activities: The Operating Company contracts waste removal to certified landfills. Equipment: From PUMP-3: Tier 1 spill kits (local) From Response Depot: 8x8 response vehicle with front-end loader/backhoe; 6x6 response vehicle with response trailer; Vacuum Truck; Skimmers; small weir and brush; Pickup truck and 4x4 automobile. Sorbents: pads and boom; Shovels, gloves, and storage bags and High-pressure washers. From another Response Depot: 2 persons and 8x8 response vehicle with roll-on, roll-off box. Scenario 3 - Pipeline Rupture KP 290, March. Situation Report Unknown loss at KP 290.0.

21

Potential size m3: 406-416 (leak), 2456-3556 (hole), 4589-6460 (rupture). Later confirmed: 6,000 m3. Sensitivity Categories: 10 - River and Stream. Otherwise open farm/grazing land. River Crossings: 150 m from the River; 5-30 m wide. River Velocity: 0.1 m; low level, no ice. Containment Site: AKIZ-08. Environmental Map Number: EM#33. Month: March. Oil Characteristics: Very slow flowing, high viscosity. Response Level: Initial Tier 1 – from PUMP-04 nearby; Tier 2 – Activated immediately and Tier 3 – Notified immediately; Equipment requested after review of the spill site. Response Objectives: Protect worker health and safety; Stop the leak; Complete government and downstream industry notifications; Prevent spread of oil into the environment, particularly downstream in the stream/river; Remove pooled oil; Remove contaminated soils and Properly dispose of wastes. Post Cleanup: Restore the area with clean material. Response Activity (by hour / days): 0-1 hours: Leak observed; Tier 1 cleanup initiated from adjacent PUMP-04; Response Contractor called out; Primary response depot (3.5 hours away) and Secondary response depot (3.9 hours away) activated and Other equipment depots initiate backfill of equipment activated equipment depots. 1-4 hours: Tier 1 equipment exhausted; Leak controlled; estimated volume lost: 6000 m3; Heavy equipment obtained from nearby government agency (from the Government Equipment Database) arrives on scene; On site observers report oil has reached stream; From Containment Site database and review of EM maps; EM#33 and EM#34; indicate that downstream sites are AKIZ-07, 06, and 03 and Satellite phone communications directs the response depot to first setup in the furthest downstream (AKIZ-03) which is directly along their travel path. After setup at AKIZ-03, if no oil, they will move up to site AKIZ-06. The response team will go to AKIZ-07, 08 and then the spill site. 4-10 hours: Response Contractors arrive; Oil is near solid floating in cold water, stream levels are low; Oil has not reached to AKIZ-07. Equipment is redeployed to AKIZ-07, 08 and the spill source and

22

Huseyin Murat Cekirge:

Response Strategy and Scenarios for Accidents in Crude Oil and Gas Pipelines

Lodging obtained in nearby city for Response Contractor workers (from Contacts Database). Day 2–4: Collection focuses on placement of filter fences, and use of viscous oil skimmers in the stream, and use of berms and dikes to stop oil movement at the source; All viscous and brush skimmers obtained from all response depots (128 m3/hr capacity); Tier 3 Contractor is instructed to supply additional skimmers, scheduled to arrive in 36 hours at nearby airport by regular air transport, airport and customs arrangements made and arrive in onscene by Day 4; Collected oil is placed in lined storage pits dug with heavy machinery; and Additional equipment contracted and obtained from government commercial agencies by using Contacts Database and Government Equipment Database. Day 5-10: Pooled oil operations completed. Focus is on removal of contaminated sediments to temporary (lined) storage pits; Specialists in bio-remediation are contracted; and A rotation plan developed for Response Contractor personnel and supervisors, taking into account additional laborers obtained locally. Day 11-30: Many Response Contractor workers return to base. Supervisors and heavy equipment operators remain; Land obtained for bioremediation; and The Operating Company contracts for site restoration. Day 30+: Cleanup terminates; Bioremediation, site restoration and monitoring begin; and Monitoring and bioremediation continue until complete. Equipment: Tier 1 spill kits (local) from PUMP-4; Heavy Equipment and Transport Vehicles - from Response Depot as listed in the Contacts Database and Government Equipment Database. From Response Depots: 8x8 response vehicle with front-end loader/backhoe; 6x6 response vehicle with response trailer; Pickup truck and 4x4 automobile; Portable storage tanks; Plastic liner for storage pits; Pumps and skimmers (weir, viscous and brush); Sorbents: pads and boom; Shovels, gloves, and storage bags; Steam cleaners; and Roll-on/roll-off boxes, accommodation module; and Communications package. From Other Depots: Viscous and brush skimmers; Filter fence kits; Replacement equipment and personnel; Additional laborers and transport obtained from nearby city, working with Government’s office. All are given appropriate

training by the Response Contractor; and Near solid oil is transport in roll-on/roll off boxes to Marine Terminal for temporary storage and treatment. Scenario 4. Large Release from Vessel Cargo Tanks, August. Situation Report: Unknown loss from vessel Marine Terminal, structural failure. Potential size m3: 1,000-10,000, duration 1-12 hours. Later confirmed: 10,000 m3. Sensitivity Categories: Marine waters; all shoreline sensitivities, recreational facilities. Water Movement: 0.5 m/sec southeast (1 knot). Containment Sites: All sites 0 to 50 km to the south: AMCS-06, 07, 08, in and near Lagoon. Environmental Map Number: EM#53 to #57 + Coastal Sensitivity Maps and Containment Manual + Nautical Chart of Area. Month: August. Oil Characteristics: Medium to light viscosities, increasing as the oil weathers. Response Level: Tier 2 – Activated immediately; Tier 3 – Notified immediately, placed on standby alert. Tier 3 – Later, activated to receive specific equipment. Response Objectives: Protect worker health and safety; Stop the leak; Complete government and downstream industry notifications; Notify Coast Guard to close the area to fishing fleet; Notify local governments to the south of potential impacts to recreational beaches; Prevent spread of oil on the water; Protect Lagoonal shorelines; Prepare for beach cleanup; Clean contaminated shorelines; Properly dispose of wastes; and Post Cleanup: Evaluate and restore impacted areas. Response Activity (by hour / days) 0-1 hours: Leak observed. Loading operations stopped. Vessel undertakes actions to stop loss from cargo tank; Response Contractor notified to activate all Response Depots; and Tier 3 contractor notified and placed on standby. A close contact with government will be established during the incident. 1-2 hours: Oil spill model trajectory shows oil movement to the south into open water, but with potential impact to south shore of Lagoon; Open-water boom deployed from the harbor and towed by workboats and line-handling boats to surround the stricken vessel; Oil Spill Response Vessel (OSRV) with onboard skimming

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 18-25

system and recovery tank departs is coming to the accident area; High-speed rubber boat deployed for oil spotting; Large capacity brush skimmer placed on tugboat; and Large capacity weir skimmer placed on another tugboat. 2-6 hours: Tier 3 contractors instructed to deliver large skimmers, open-water boom, dracones for on-water storage, and transfer pumps, delivery is cleared for Airport in 24 hours, expected onscene in 30 hours; The Response Contractor will handle local transport and expedite delivery, Customs Agency notified of incoming delivery, Tier 3 contractor supplies list of additional operators to be hired to operate the equipment; Booming of the vessel completed, with approximately 30% of oil being retained; On-water boom with the workboats successfully encircle large oil patches; Using Contacts Database, large barge is contracted for on-water storage, arrival in 24 hours; Oil recovery begins on water with the OSRV; Storage barge units (35 m3) deployed in the water and pushed to work area by workboats; OSRV is filled, and transfer begins to barge units. Barge units bring back oily water to small boat harbor; and Vacuum trucks transfer barge cargo to the ballast water treatment. 6-24 hours: Oil remains at sea; On-water recovery operations continue at sea using two skimmer boats and three outfitted tugboats with storage; Storage barges and tugs make 20 km roundtrip for offloading at the existing boat harbor; Boom deployed at the designated Containment Sites; On-water operations return at darkness, after risk analysis, pumping operations are allowed to continue at the Terminal under portable lighting; and A night shift mans the Incident Command Center, planning for the next day’s activities. Day 2: A wind shift pushes oil to south, sand-dominated shoreline of Lagoon; model results indicate impacts late on day 2, Offshore operations continue using the two OSRVs, three tugboats, and four on-water storage barges; Training of field crews for shoreline cleanup is completed; Field teams from other Response Depots are sent to sand beaches south of to prepare for impacts by removing debris from the waterline; Using Government Equipment Database, Government Agency agrees to send heavy equipment (road graders and front-end loaders to assist sand beach cleanup); Oil impacts start on shoreline and into eroding marsh area of south side of Lagoon; Shoreline cleanup concentrates on oil removal from the high-tide strand line using mainly manual labor working with the front-end loaders avoiding use of heavy machinery in oiled areas. As the tide drops, road graders effectively sweep oil and

23

minimal sand into rows for removal by front-end loader. Sites are designated for the temporary storage of oily sand; and At end of day, additional skimmers and equipment from Tier 3 contractor arrives, large barge storage also onsite. Day 3: On-water recovery is re-deployed taking advantage of the extra skimmers and storage capacity; Shoreline cleanup continues, being directed by data provided by the Shoreline Cleanup Assessment Team (SCAT); Marsh cleanup will start only after impacts end; Oily sand is transported to a landsite for potential bioremediation of disposal; Oil spill modeling indicates a likelihood of further impacts far to the southeast; and Response teams are directed to activate containment sites AMCS-03, 04 and 05. Day 4; On-water operations recover decreasing amounts of oil, Wave conditions send most units to shelter; New shoreline impacts occur near in the area; Beach cleanup is divided into two groups: south of Lagoon, and in the nearby area, the latter concentrates on oil removal from recreational beaches; and The Response Contractor supervises shoreline cleanup operations using newly trained workers, other staff and unneeded equipment return to their respective depot, marsh cleanup focused on removal of surface oil by hand laborers begins. Day 5 – 10: On-water operations are completed; Decontamination of equipment begins, Tier 3 material is returned to base; Shoreline cleanup operations completed in the northern work area, a monitoring program of the area is developed. Workers shift to the south; and Work in the south completes oil removal from sand beaches. Work on structures and bed rock area continues, contaminated sand is transported to an inland site for temporary storage or bioremediation, other contaminated material (sorbents, etc.) is sent to an approved landfills. Day 11-30: A termination of cleanup operations is approved with a monitoring program in place, and with the standby of additional response measures as needed; A bioremediation plan is approved for treatment of oily sands, with a ground water monitoring program in place; Response material is decontaminated or disposed; and A SCAT assessment indicates restoration work is needed in recreational areas and in oiled marshes. Day 30-45: SCAT surveys continue on a weekly basis; Response teams respond to additional intermittent oiling; and Restoration work is undertaken by a hired contractor. Day 45-60: Operations terminate.

24

Huseyin Murat Cekirge:

Response Strategy and Scenarios for Accidents in Crude Oil and Gas Pipelines

Equipment: All equipment from the Response Depot; From other depots: fast flow and other skimmers, pressure washers, vehicles, storage boxes, vacuum trucks, PPE, etc.; and Personnel from all response depots, keeping a basic staff in each response depot for abling to effectively respond to a new spill in their area. Outside support: Heavy equipment from government agencies, storage barge from private supplier, Tier 3 skimmers, storage and boom. Purchased equipment: additional spray and pressure washers, PPE, work clothes, storage bags, etc.; Terminal support: workboats, line-handling boats, tugboats, oil-water separator and oily water transport; Additional Laborers are given training and hired for shoreline cleanup facility at the existing terminal; Tugboats setup two large skimmer systems around the terminal, and use pumps to lift oil to oily waste on the terminal for direct hard-line transfer to existing oil-water separator (capacity: 160 m3/hr); After setup, tugboats redeploy for on water recovery using smaller disk and fast flow skimmers (capacity up to 30 m3/hr); and Boom and small skimmer systems sent to Containment Sites AMCS-07 and AMCS-08, incoming response team from other equipment depot is directed to cover AMCS-06 (deep within Lagoon). 2.2. Response Scenarios for NG and NGL Pipelines A sample scenario is the following: Explosion at KP 220 is reported is reported. Scenario - Explosion and Fire at KP 220, September Situation Report Explosion at KP 220 is reported. Location: KP 220 Month: September. Gas Characteristics: Natural Gas. Response Level: All levels. Response Objectives: Protect worker health and safety; Stop the leak ; Stop the fire; and Rescue people if they are influence by fire, explosion and toxication. Response Activity (by Hour): 0-1 hours: Leak controlled; Influenced area and communities are determined through modeling techniques; Fire extinguishing starts; Later hours and days: Fire extinguishing continues; Help for toxicated people continue; and

Wreckage cleanup continues. Additional activities: The Operating Company contracts waste removal to certified landfill. Equipment: Fire engines; Wreckage moving heavy equipment; PPE and Cleaning equipment

3. Conclusions The paper presents an important issue of actions during the pipeline accidents. The material presented here is guidance for preparing planning, training and real time actions program for disaster situations. It should mentioned that each pipeline and its terrain have own particular specifications. The appropriate scenarios must prepare before the pipeline starts the operations. For optimum response, priory scenarios help the decision makers; it should be noted that probability of occurrence of a priory scenario is quite remote. However, the practice on priory scenarios will lead better actions by the operating company’s management.

Acknowledgements Thanks to Cihan Anul for discussions during progress of the paper.

References [1]

Huseyin Murat Cekirge, Outlines of an Oil Spill Response Plan (OSRP) for Crude Oil Pipelines, International Journal of Environmental Monitoring and Analysis. Vol. 3, No. 3, 2015, pp. 191-197. doi: 10.11648/j.ijema.20150303.21

[2]

Huseyin Murat Cekirge, Omar K. M. Ouda, Ammar Elhassan, A Method for Preparing Environmental Social Impact Assessment (ESIA) of Crude Oil and Gas Pipelines, International Journal of Environmental Monitoring and Analysis. Vol. 3, No. 3, 2015, pp. 154-161. doi: 10.11648/j.ijema.20150303.17

[3]

Huseyin Murat Cekirge, Environmental Risk Assessment Methodology (ERAM) for Oil Pollution, International Journal of Environmental Monitoring and Analysis. Vol. 3, No. 2, 2015, pp. 91-110. doi: 10.11648/j.ijema.20150302.18

[4]

Huseyin Murat Cekirge, Quantitative Risk Assessment for Crude Oil Pipelines, International Journal of Environmental Monitoring and Analysis. Vol. 3, No. 3, 2015, pp. 147-153. doi: 10.11648/j.ijema.20150303.16

[5]

W. J. Lehr, R. J. Fraga, M. S. Belen, and H. M. Cekirge, "A New Technique to Estimate Initial Spill Size Using a Modified Fay-Type Spreading Formula," Marine Pollution Bulletin, 1 5(9), 326-329, 1984.

[6]

W. J. Lehr, H. M. Cekirge, R. J. Fraga and M. S. Belen, "Empirical Studies of the Spreading of Oil Spills”, Oil and Petrochemical Pollution, 2 (1), 7–11, 1984.

[7]

H. M. Cekirge, “Oil Spills: Determination of Oil Spill Volumes Observed on Water Surfaces”, The International Journal of Technology, Knowledge And Society, 8 (6), 17 – 30, 2013.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 18-25

[8]

Huseyin Murat Cekirge, Qualitative Risk of Gas Pipelines, American Journal of Energy Engineering. Vol. 3, No. 3, 2015, pp. 53-56. doi: 10.11648/j.ajee.20150303.14.

[9]

FEMA, Emergency Response Guidebook, 2008.

[10] FEMA, DOT, and EPA, Handbook of Chemical Hazards Analysis Procedures, US Environmental Protection Agency, US Department of Transportation, and the Federal Emergency Management Agency, 1989. [11] United States Environmental Protection Agency (EPA), Evaluation of Dense Gas Simulation Models, Draft, EPA 450/4-90, United States Environmental Protection Agency, September 1990. [12] United States Environmental Protection Agency (EPA), Risk Management Program Guidance for Offsite Consequence Analysis (OCAG), EPA 550-B-99-009, United States Environmental Protection Agency, April 1999.

25

Special Programs Administration, 49 CFR Part 192, [Docket No. RSPA00-7666; Amendment 192-95] RIN 2137-AD54, Pipeline Safety: Pipeline Integrity Management in High Consequence Areas (Gas Transmission Pipelines), ACTION: Final rule. [17] U.S. Department of Commerce (US DOC), "Heat radiation from Large Pool Fires" NISTIR 6546, Fire Safety Engineering Division Building and Fire Research Laboratory, November 2000. Risk Assessment, American Institute of Chemical Engineers, New York, 1995. [18] W. K, Muhlbauer, Pipeline Risk Management Manual, Second Edition, Gulf Publishing Co., Houston, TX, 1996. [19] W. K, Muhlbauer, Pipeline Risk Management Manual, Third Edition, Gulf Publishing Co., Houston, TX, 2004.

[14] United States Department of Transportation Office of Pipeline Safety (DOT-OPS), Gas Pipeline Incident Database, 1984-200 1, and Hazardous Liquid Incident Database, 1986-2001.

[20] United States Environmental Protection Agency (EPA), ALOHA®Areal Locations of Hazardous Atmospheres, User's Manual, United States Environmental Protection Agency, Office of Emergency Management, Washington, D.C. and National Oceanic and Atmospheric Administration, Office of Response, Seattle, Washington, February 2006. United States Environmental Protection Agency (EPA), Download of ALOHA, Version 5.4, http:www.epa.Rov/ceppo/cameo/request.htm. Accessed September 2006.

[15] United States Department of Transportation Office of Pipeline Safety (DOT-OPS), 2000 Annual Reports for Natural Gas Transmission and Distribution Pipeline Operators.

[21] Mark W. Eltgroth, "CHARM® Emergency Response System Technical Reference Manual," Radian Corporation, September 1995.

[16] United States Department of Transportation, Research and

[22] Radian International, CHARM® Technical Manual, 1995.

[13] United States Environmental Protection Agency (EPA), TANKS 4.09b, Computer Program, US Environmental Protection Agency, September 1999.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38 Published online October 15, 2015 (http://www.sciencepublishinggroup.com/j/ijema) doi: 10.11648/j.ijema.s.2015030601.14 ISSN: 2328-7659 (Print); ISSN: 2328-7667 (Online)

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines Huseyin Murat Cekirge Department of Mechanical Engineering, the Grove School of Engineering, the City College of the City University of New York, New York, USA

Email address: [email protected]

To cite this article: Huseyin Murat Cekirge. Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines. International Journal of Environmental Monitoring and Analysis. Special Issue: Environmental Social Impact Assessment (ESIA) and Risk Assessment of Crude Oil and Gas Pipelines.Vol. 3, No. 6-1, 2015, pp. 26-38. doi: 10.11648/j.ijema.s.2015030601.14

Abstract: Containment sites are the most important locations for recovering spilled oil on land and marine environments. Analyses for determining of containment sites for crude oil pipelines are explained in detail. Necessary information for designing land and marine containment sites is introduced. The necessary databases for efficient response are also mentioned. The paper is guidance for planners who will design oil spill response plans. Keywords: Land Containment Sites, Marine Containment Sites, Catchment Areas, Land Oil Spills, Oil Spill Response Plans, OSRP Databases

1. Introduction A forecasting and trajectory model is necessary to follow and monitor the oil on land and sea. This program may help to resolve legal claims aftermath of the oil spill. Monitoring oil spills is; Spill size determination; Spill movements; Spill tracking that help Incident Command; and Environmental monitoring. Oil spill can be initiated from oil tanks, terminal, open water, vessels and pipelines on land. OSRP (Oil Spill Response Plan) must consist of detailed information for each case, [1 - 6]. Based on the size and location of the whole installation, equipment depots should be established for proper response; even considering air transport for heavy equipment. The locations must be chosen for the most efficient response to onshore and offshore oil spills. Land Spills: On land, oil spills start from pipelines and flow low spots by gravity and topographic contours through catchment areas and end up river beds, Figure 1. At the each kilometer point, an oil spill may start and reach a river bed, which is called “containment site”. It is mostly required two containment site for every Kilometer Point, KP, in other words for a segment between two consecutive KP. The cleaning operations can performed at these containment sites. Each containment site must be analyzed; and the

following information must be recorded in pictures, maps and sketches: Description of the site and directions; General cleanup options / equipment; Environmental sensitivity of the pipeline corridor; Pipeline crossings information; and Downstream environments.

Figure 1. Down slope potential oil spill flow analysis (Catchment Area), related containment sites and pipeline KPs (Kilometer Points), [1].

The spills must be caught and cleaned from the equipment coming from equipment depots which are

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

established best reachable location for containment sites, even considering Tier 1, Tier 2, and Tier 3, [1]. Oil will flow toward low spots under the influence of gravity. In general, it will follow gullies and dry creek beds at right angles to topographic contours toward surface water. At colder temperatures, crude oil in the pipeline system will become extremely viscous and may reach its pour point, below which it will be essentially immobile. The probable pathway of spill movement has been determined along the entire pipeline by analysis of topographic maps supported by fieldwork. The degree of threat to groundwater presented by a petroleum hydrocarbon spill is subject to variables including the properties of the hydrocarbon, size and location of the release, permeability of the soil impacted, depth to groundwater, and effectiveness of any response action. Petroleum hydrocarbons will adsorb onto soil particles and be held in soil pore spaces by capillary action. Left uncontrolled, a fluid hydrocarbon spill will penetrate into a porous soil until it is absorbed and bound by soil particles or until it reaches an impermeable layer or groundwater. Soils typically can retain 15 to 40 liters of petroleum per cubic meter. If temperatures are below the pour point of the hydrocarbon, threat to groundwater is minimal. Spills to Sediment Dominated Shorelines: Oil can become incorporated into beaches by burial (shifting sand) or penetration. Generally, the larger grain size results deeper potential penetration. At the extreme, it can be expected that penetration into coarse gravel/cobble beaches will be greater than on compacted sand or mud flats. On sand beaches, emulsions and waxy oils will not penetrate as readily as lighter oils and crudes. As a general rule, the maximum level of oil in sediment beaches is 10 percent by volume (where oil may be dripping out of the sediment). It should be mentioned that common levels are 2 to 5. Creeks and Streams: Oil entering creeks and streams will move downstream with the current. Maps prepared for the Containment Manual Database indicate estimated flow direction information and can be used to initially predict spill flow directions at any location along the pipeline. Some spreading may occur, particularly with diesel spills, but the spreading motion will typically be overcome by even a slight current. For planning purposes, assume that the oil will move at stream flow velocity without wind or other effects. In practice, oil will tend to accumulate in areas of quiet water, eddies, in vegetation, and in debris accumulations. Natural collection areas can frequently be identified by the presence of accumulated debris. Flow velocity can be estimated by timing the movement of a floating object, such as a stick, over a measured distance. Field measurement of ambient current conditions provides a useful method of identifying locations where booming can be successfully conducted. Storm Drains / Culverts: Storm drains in the region may include buried pipes, open box culverts and lined or unlined channels. Oil entering storm drains may be lost from view and may be difficult and dangerous to recover. These features are confined spaces where dangerous concentrations

27

of vapors may accumulate. As such, spills to storm drains can present significant fire and explosion hazard (especially for petrol spills) and constitute high-risk entry situations. Spills to storm drains may be assessed by carefully opening the nearest manhole cover and visually determining the direction of flow. If oil is present, checking observation down gradient continues until oil is disappeared. Vapor monitoring instrumentation may be used for tracing fluids and vapors. Oxygen deficient and explosive atmospheres may exist with all releases involving confined spaces. All releases to storm drains should be approached with caution. Canals and Irrigation Channels: The pipeline may lie in proximity to many water supply canals and irrigation channels. Most of these canals have elevated banks and are exposed to spills only at points where the pipeline crosses them. Other water supply canals are located along major rivers, and generally have control structures which may be used to control floating oil. Flow velocities in canals can be estimated by timing the movement of floating debris over a known distance. The purpose of this paper is to provide environmental and containment site information needed to respond efficiently during an oil spill incident generated by the operating company related activities. Information is provided on: 1. Ecological resources in the area and their seasonality; 2. The type of shorelines present and their general sensitivity to spilled oil; 3. Areas of human-use importance, particularly related to recreational use and fishing; and 4. Preplanned sites for the deployment of oil spill equipment, referred to as containment sites.

2. Land Oil Spills For the pipeline, volume estimates can be calculated by adding the estimated volume lost by pumping and drainage of material free to gravity flow from the pipeline. Loss Due To Pumping: Loss due to pumping can be calculated by multiplying the pumping rate by the time elapsed from event discovery to pump shutdown. Specific data should be available from the pipeline control center. Loss Due To Drainage: After pumping has been terminated, sections of pipeline topographically above a point of failure will attempt to drain. Adjacent sections of the pipeline may be above or below the leak point and drainage may occur from one or both directions, depending on location. In addition, pipeline valves will influence pipeline drainage. Block valves have been placed at strategic locations to control the magnitude of potential spills by isolating pipeline segments and controlling the flow of oil in both directions. Maximum pipeline loss may be complicated due to topography and the placement of block valve. To roughly determine the potential pipeline drainage, the contents of a pipeline from the topographic high down to the break point

28

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

can be used. Block valves will reduce the potential flow if between the high point and the break point. Siphoning effects over topographic highs may occur, but are difficult to estimate. The amount of oil contained in a contaminated land area may be estimated by using the following techniques: Classify the area: Land spills are usually irregular in shape. To begin with estimation of volume, classify the exposed area into either “flow” areas or “pooling” areas. Flow areas are those over which the oil moves, and typically contain an oil coating with minimal penetration. Pooling areas consist of depressions or low spots where oil accumulates. Soil penetration may occur in these areas. Estimate the area covered: To develop an estimate of the area covered by flow exposure or pooling, break each category into a series of rectangle or squares which approximate its shape. For edges, the rectangles should contain approximately fifty percent oiled surface and fifty percent unoiled surface. Multiply the length by the width of each rectangle and sum the totals to obtain the area estimate. The more rectangles you use, the more accurate your estimate will be. Estimate the Average Depth of the Oil: Oil will typically vary from very thin at the edges to thicknesses depending on the depth at the low points. Depths can be estimated by gauging with a stick at a number of locations. If the oil pool is wide or inaccessible, estimate probable depth by projecting the general surface profile to the center of the pool. Calculate the Free Oil Volume: Multiply the estimated areas by the estimated average depths to obtain free oil spill volume. Roughly 1 cubic meter of oil is 1 ton and equal to 7 barrels. Estimate Penetration into Soil: Oil may penetrate into soils. The depth of penetration is determined by factors including soil type and porosity, viscosity of the oil, temperature, and presence of water. Field observations should include sufficient small excavations to determine whether soil penetration has occurred and if it has, how deep. If penetration has occurred, estimate the depth by excavation. It should be checked to determine whether oil in the soil is mobile by allowing a contaminated sample to drain. Any oil which will drain is free to migrate vertically or horizontally through the soil. Penetration estimates are necessary for assessing the impact and identifying appropriate response actions. Burrows: In some areas spilled oil may enter animal burrows. Depending on its characteristics, oil entering burrows may remain in the burrow, or soak into surrounding soils. Oil hidden in burrows is often a source of long term recontamination. Estimate Penetration into Snow: In cold temperatures below its pour point, the pipeline system crude oil into snow is anticipated. However, other hydrocarbons may exhibit the ability to soak into snow. When evaluating

potential contamination in snow, always consider the fact that spillage may be hidden from view by recent snowfall. Small observation trenches may be necessary to determine presence and extent of contamination in such cases. Always plot the location of any contamination on snow so it can be re-located after additional snowfall. It is recommended that this be done by walking the perimeter of the spill area using a handheld GPS unit and downloading the resulting track line. It is also advisable to stake locations around the perimeter of the spill. This oil must be recovered right before snow in the start to melt. Tank-related spills: These include spills from above ground storage tanks and tanker trucks. The volume of tank spills may be estimated by comparing pre-spill fluid levels with post-spill levels or with the height of the hole, assuming the tank dimensions or capacity is known. 2.1. Containment Sites The containment sites are the most important for mitigation of oil spills and its effects on environment. These sites must be determined as the end of the catchment areas, every one kilometer segment should be assigned at least two containment sites for the collection of spilled oil. The equipment depot must be at distance in which the response must be started according to the operating company’s policy. These containment sites must be regularly inspected and the pertinent data must be collected. The “Containment Site Database” must be designed and updated. There other response databases are: Equipment database: The database for locating available equipment could be used during oil spills. This database is designed for obtaining equipment from the local and international resources. Notification Database: The database for notifying authorities and interested parties who can help response activities. These parties could be governmental, private, medical and legal organizations that will facilitate the response activities and accommodation of the response staff. It is again, the databases must be updated regularly and tested for its functionality. Mammal database: The database contains in mammals around the pipeline. There may be some other databases related to transported crude oil and environmental conditions of the pipeline location. The data collected for the containment site are as follow: Location; Pipeline information; Necessary and proper for equipment for the site; Sensitivity of the area and Pertinent topographical environmental maps; and Pictures. These data can be seen in Figures 2, 3, 4 and 5. Equipment list for the equipment depots can be prepared after the analysis and determination of appropriate equipment for the the containment sites.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

2.2. Tracking Oil in Groundwater and Karst Areas In areas where groundwater has been contaminated, wells adjacent to the contamination site will be monitored for oil contamination. Professional scientists will set up the monitoring and instrumentation program working with the government

29

water agency, and related organizations. In karst areas, and in areas with few or no operating water wells, new wells must be drilled and pumped to determine the extent of subsurface contamination. This is a highly specialized field. Personnel from local groundwater organizations will be consulted to oversee and/or advise on such operations.

Figure 2. The location of the containment site, [7].

30

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

Figure 3. The sensitivity and proper equipment of the containment site, [7].

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

Figure 4. The environmental properties of the containment site, [7].

31

32

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

Figure 5. Pictures of the containment site, [7].

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

2.3. Monitoring Guidelines The basis of an environmental monitoring program is to detect change in selected parameters over time. It will be carried out in two parts: Comparisons of post-spill data with pre-spill data (a baseline); Comparisons of post-spill data from the polluted area with data from an unpolluted, representative reference or control area; and The baseline identifies conditions before the incident. The reference or control area provides information on natural site-to-site and seasonal fluctuations. The design of a monitoring program will focus on objectives and specific measures to be used, i.e. 'endpoints'. For most credible scientific studies, a testable hypothesis will be formulated. For example, Objective: Does manual clean-up affect the recovery rate of oiled areas? ; Hypothesis: Manual clean-up of areas does not influence the long term recovery compared to no treatment; and Endpoint (Measure): Distribution and percentage cover of flora by oil within manually cleaned, uncleaned and unpolluted areas. The basis of monitoring is to identify changes through time. However, if change is identified, the cause of change is not always directly attributable to the oil spill that prompted the study. There can be large natural fluctuations in community structures and species populations. The use of reference areas would serve to identify these and prevent them being attributed to an environmental stress such as an oil spill. When formulating monitoring objectives, any relevant constraints should be recognized and their impacts taken into account. The specifics of a monitoring program are dependent upon the details of any given incident. Among factors to consider are: Identification of relevant baseline information; Inclusion of proper and useful reference or control sites in the study; Establishment of objectives of the monitoring program, including endpoints; Establishment of sampling methodologies; for example quantitative versus qualitative measurements; What constraints are being applied and what mitigations can be brought to carried on; and External reviews, such as peer review of the study and results and engagement of the general public.

3. Marine Oil Spills As appropriate, the operating company will establish a spill tracking and monitoring program to assist in assessing the conditions of the affected marine and the continuing effectiveness of the response activities. This program may also assist resolving damage claims. Close liaison will be

33

maintained with communities affected by the spill to ensure their concerns are addressed. The program may be relatively simple for small spills show little impact, or very complex for large spills or spills affecting particularly complex or sensitive habitats. The implementation of a well-planned monitoring program facilitates a greater understanding of the specific impacts of an incident, as well as improving overall knowledge for future incidents. Key elements of the spill tracking and monitoring program include: Spill Size Assessment – Estimating spill volumes; Spill Movement – Estimating where oil will go; Spill Tracking - Monitoring of the characteristics and movement of released oil will be conducted periodically during the event to provide real-time information to Incident Command. Shoreline Cleanup Assessment Teams (SCAT) may be required to document the extent and distribution of shoreline oiling; and Environmental Monitoring – Involving regular monitoring after an incident up to the point where agreement is reached that remediation is complete, thereby providing information on the progress of recovery in the area. The operating company will provide oversight and work with the Response Contractor to provide documentation, observations and sampling of the ephemeral and longer term environmental impacts of the oil spill and associated cleanup operations, striving in all cases to reduce the short and long impacts of these operations. 3.1. Spill Size Assessment Terminal: A preliminary estimate of the size of a pipeline spill is necessary to gage the size of the response. It can be made from the SCADA system, from calculating pumping loss from the time of the incident until shutdown, from estimates of static pipeline drainage, and from visual estimates. It is important to treat calculations based on visual observations as preliminary, and for response planning purposes only. The magnitude of the release may change with time. In any case, precise calculation of the volume of a release is difficult and may not be possible until the pipeline is repaired. Release figures may be used to set fines and assess damages. Vessel Related: Losses from tanker vessels will primarily be related to spills from the cargo or fuel tanks. Before and after measurements are most commonly used to determine the actual losses. If a rupture occurs during a loading operation, then spill size can be calculated based on hole size and location of the leak, and estimated leak duration. The location of the hole with respect to the waterline and interior fill line is important. The pressure head inside the cargo tank must be greater than the outside pressure for material to exit. The amount exiting a tank will eventually reach equilibrium with the outside water pressure if the leak is below the water line. These calculations can be difficult and sufficient information is typically not available during the first stages of a spill.

34

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

Open Water: Once oil is on the water, accurate visual estimates are difficult to achieve because spilled oil is seldom uniformly distributed. The basic procedure involves estimation or measurement of the spill surface area, multiplied by its estimated average thickness of the oil. The thickness of oil on water can be estimated from its color, [8 - 10]. Emulsions: In some cases, mechanical energy may have mixed water into spilled crude oil, forming an emulsion, or mousse. A practical measure of emulsion stability and percentage can be obtained by allowing an oil sample to sit in a glass container and after 12 to 24 hours, measure the water fraction (water layer, if any). Refined products are not expected to demonstrate significant emulsification. 3.2. Estimating Spill Movement Open Water - Using a Computer Model: An oil spill fate and trajectory model must be available for use at marine terminal. Utilizing actual winds and concurrently measured tides, the model provides an estimate of spill movement under the actual conditions present at the time of the spill. The procedures for model use must be developed. Open Water - Vector Addition: Spilled oil movement on water can be estimated graphically by adding wind and current vectors. Surface currents will dominate spill movements unless the winds are extremely strong. Observations in actual spill situations have shown that wind will cause an oil slick to move at about 3% of the wind speed, and in the same general direction. 3.3. Tracking and Surveillance The design of a monitoring program will focus on objectives and specific measures to be used, i.e. 'endpoints'. For most credible scientific studies, a testable hypothesis will be formulated to investigate affects of clean-up on the recovery rate. A key element in an effective spill response is knowledge of where the spilled oil is located and tracking its movements. The environment unit will work to provide the appropriate level of oil spill tracking required by Incident Command. This section describes procedures for effectively tracking oil when: On the water’s surface (rivers, lakes and oceans); On shorelines; Below the water’s surface (i.e. sunken oil) and Underground movement via groundwater or in karst areas. In all cases, the location of the observation is important. To this end, 1:30,000 detailed topographic maps with UTM (Universal Transverse Mercator) coordinates must be prepared for the entire route and areas of potential marine impacts. The specific location of the observation must be determined using a GPS (Geographic Positioning System). Photographs and video imaging can be marked and/or coordinated to GPS location by using software and titling units. Suppliers able to provide this expertise during oil spills are listed in the Contacts Database. Aerial surveillance via helicopter is the fastest and best

method to track oil on the water’s surface where long distances are involved, and also offshore. Fixed-wing or twin-engine aircraft may be a required alternative. The Contacts Database provides information to obtain a helicopter or fixed-wing from government and private sources. General rules for aerial surveillance are as follows: Trained or experienced observers should be utilized; Communication to the pilot must be maintained via headsets and hand signals; The observer should sit in front with the pilot; The ability to view through an open window is preferred; and Low altitude (200 m), slow flight is preferred, with the necessity of going up or down, or landing, for oil verification. During the initial spill stages, frequent overflights (two or more times a day) are necessary: Detailed (1:30,000) maps must be brought on board to mark the flight path and mark the distribution of oil; Oil is likely to be patchy (for example, large areas of no oiling followed again by oiled shorelines or offshore patches are common) and The NOAA (National Oceanic and Atmospheric Administration) Open Water Oil Identification Job Aid (http://response.restoration.noaa.gov/shor_aid/shor_aid. html), [11], may be used, providing: Checklists; Photo examples of oil (and oil-similar material) on the water’s surface; Example overflight maps; and Glossaries and charts. Tracking Oil on Shoreline: Shoreline surveillance is carried out by a combination of aerial and ground surveys with the purpose of determining oil concentrations on the shoreline to advise Operations for cleanup. This is a highly specialized activity, and these activities must be done correctly. The Contacts Database lists companies available to provide this service, [12- 16]. Tracking Oil below the Water’s Surface: Highly weathered or bunker oils, or oils combined with sediment, may sink below the surface of the water, presenting a much more difficult problem in tracking the distribution of the spill. Methods to determine oil in the water column or on the bottom include: Small fishing nets, partially lined with sorbents, to verify oil presence (black spots on the white sorbent); Sorbent pads attached to a weight or anchor, and dropped to the bottom; Bottom dredges (used to collect shellfish); and Oceanographic benthic (bottom) survey instruments (grab sampler). A portable flow-through fluorimeter may also used to track dissolved oil in the water column, but involves sophisticated instrumentation.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

3.4. Onshore Containment Sites Spilled oil finally moves towards to shorelines and threatens environmentally sensitive areas; and these areas are estuaries, lagoons and beaches. The oil spilled from pipelines and pipeline’s operations threatens marine areas of the pipeline terminal. The sensitive areas and containment sites can be determined by simulating oil spills for various environmental and meteorological conditions. Once, these sensitive areas areas determined, these are considered like land containment sites for collecting spilled oil in marine environment. These areas must be inspected, and the pertinent data of the area are collected for establishing a database. The

35

rules of marine containment sites are same as the land containments sites. The difference is the activation process, which is determination of the containment sites that will be threatened by the actual oil spills in the area. After activations process, all the actions for recovering oil in containment sites will be started. The equipment for each containment site must be selected; and then the equipment inventory for marine spills of response depots can be determined. These equipments will be in the general equipment database which will be used together with notification and other databases. The details of the pertinent data are presented by Figures 6, 7 and 8.

Figure 6. Onshore containment site, [17].

36

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

Figure 7. Activation conditions and appropriate equipment for the containment site, [17].

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 26-38

Figure 8. Pictures and sensivity of the containment site, [17].

37

38

Huseyin Murat Cekirge:

Determination of Land and Marine Containment Sites of Oil Spills from Crude Oil Pipelines

[7]

H. M. Cekirge, Pipelines, Maltepe Uni., Int. Rep. 1/2, Istanbul, 2010.

[8]

H. M. Cekirge, Oil spills and shorelines, Maltepe Uni., Int. Rep. Cekirge, H. M., “Oil Spills: Determination of Oil Spill Volumes Observed on Water Surfaces”, The International Journal of Technology, Knowledge And Society, 8 (6), 17 – 30, 2013.

[9]

W. J. Lehr, R. J. Fraga, M. S. Belen, and H. M. Cekirge, "A New Technique to Estimate Initial Spill Size Using a Modified Fay-Type Spreading Formula," Marine Pollution Bulletin, 1 5(9), 326-329, 1984.

4. Conclusions The paper presents detailed analysis of determination of land and marine containment sites for crude oil pipelines. The data must be collected to design these containment sites, and the related databases must be prepared and regularly updated. The procedure of determination of these sites and the required data are explained, and practical examples are given including pictures. The methodology is the directions for preparing oil spill response plans.

References [1]

[2]

[3]

[4]

Huseyin Murat Cekirge. Outlines of an Oil Spill Response Plan (OSRP) for Crude Oil Pipelines. International Journal of Environmental Monitoring and Analysis. Vol. 3, No. 3, 2015, pp. 191-197. doi: 10.11648/j.ijema.20150303.21. Z. Zhong Z and Y. You, Oil spill response planning with consideration of physicochemical evolution of the oil slick: A multiobjective optimization approach, ANL/MCS-P1786-0810, Mathematics and Computer Science Division, Argonne National Laboratory, 2010. D. Mackay, I. A. Buis, R. Mascarenhas R and S. Paterson, Oil spill processes and models: Environment Canada, Manuscript Report No 8. EE-8; Ottawa, Ontario, 1980. N. P. Ventikos, E. Vergetis, H. N. Psaraftis and G.Triantafyllou, A high-level synthesis of oil spill response equipment and countermeasures. Journal of Hazardous Materials; 107:51-58, 2004.

[5]

R. G. Pond, D. V. Aurand and J. A. Kraly, Ecological risk assessment principles applied to oil spill response planning in the Galveston Bay Area, Texas General Land Office, Austin, Texas, 2000.

[6]

Z. Zhong and F. You, Oil spill response planning with consideration of physicochemical evolution of the oil slick: A multiobjective optimization approach, Computers & Chemical Engineering, 08/2011; 35(8):1614-1630. DOI: 10.1016/j.compchemeng.2011.01.009, 2011.

[10] W. J. Lehr, H. M. Cekirge, R. J. Fraga and M. S. Belen, "Empirical Studies of the Spreading of Oil Spills”, Oil and Petrochemical Pollution, 2 (1),7–11, 1984. [11] NOAA (National Oceanic and Atmospheric Administration) Open Water Oil Identification Job Aid (http://response.restoration.noaa.gov/shor_aid/shor_aid.html), 2010. [12] NOAA (USA, National Oceanic and Atmospheric Administration) Shoreline Assessment Job Aid: (http://response.restoration.noaa.gov/shor_aid/shor_aid.html), 2010. [13] Environment Canada, (http://www.etc-cte.ec.gc.ca/estd_west/estdwest_scat_e.html# 02), 2010. [14] NOAA (National Oceanic and Atmospheric Administration), ( http://response.restoration.noaa.gov/oilaids/shore/shore.html), 2012. [15] Environment Canada, (http://www.etc-cte.ec.gc.ca/estd_west/SOS%20FORM.pdf.), 2010. [16] NOAA (National Oceanic and Atmospheric Administration), Assessment of the Risks Associated with the Shipment and Transfer of Group V Fuel Oils (NOAA, 1994), (http://response.restoration.noaa.gov/oilaids/GroupV.pdf), 1994. [17] H. M. Cekirge, Oil spills and shorelines, Maltepe Uni., Int. Rep. 1/3, Istanbul, 2010.

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 39-46 Published online October 15, 2015 (http://www.sciencepublishinggroup.com/j/ijema) doi: 10.11648/j.ijema.s.2015030601.15 ISSN: 2328-7659 (Print); ISSN: 2328-7667 (Online)

Determination of Risk to Groundwater Aquifers from Crude Oil Pipelines Huseyin Murat Cekirge Department of Mechanical Engineering, the Grove School of Engineering, the City College of the City University of New York, New York, USA

Email address: [email protected]

To cite this article: Huseyin Murat Cekirge. Determination of Risk to Groundwater Aquifers from Crude Oil Pipelines. International Journal of Environmental Monitoring and Analysis. Special Issue: Environmental Social Impact Assessment (ESIA) and Risk Assessment of Crude Oil and Gas Pipelines.Vol. 3, No. 6-1, 2015, pp. 39-46. doi: 10.11648/j.ijema.s.2015030601.15

Abstract: The pipelines are passing over groundwater aquifers and any oil spills from the pipeline is major threat to the aquifers. The methodology to determine this risk will be introduced; the detailed analysis will be explained and an example is presented. The software used in the calculation is also explained. Keywords: Crude Oil Pipeline’s Risk to Groundwater, Zone of Contribution, Capture Zones, Risk Criteria for Groundwater from Crude Oil Pipelines

1. Introduction Required elements of the study include data collection of well sites and each aquifer, and use of a computer-based groundwater flow model to determine 50-day and 400-day capture zones and the Zone of Contribution (ZOC), [1 - 11]. The analyses are required to be updated for every kilometer of pipeline (KP, kilometer point) crossing the aquifer as necessary; and also to reflect the results obtained. The methodology must be developed across the entire pipeline to provide guidance for environmental protection. Lastly, oil spill response procedures will be developed based on the knowledge gained during this process. During this analysis, a steady state geohydrology computer program called Wellhead Analytical Element Model (WhAEM) 2000 or WhAEM, [2], developed by the United States Environmental Protection Agency, is used in the evaluations. Well data will be extensively updated with a database, map information, reports and well logs. Model runs were made to determine the 50-day, 400-day and 4000-day capture zones of wells in each aquifer along the pipeline. The wells in the sample aquifer are used only for seasonal irrigation, so the model run was made for 300-days which represent the maximum number of operational days within a 400-day period. This risk assessment procedure, the Groundwater occurrence-Overall lithology-Depth to groundwater (GOD) method is described in the World Bank website publication; Assessment of Groundwater Pollution

Risk; Morris and Foster, [4]. Taking the ‘conservative’ approach, the Quantified Risk Assessment (QRA) risk criteria were used to develop groundwater risk values for every kilometer; KP, of pipeline crossing each aquifer. Data from this study are incorporated into the Project’s Geographic Information System (GIS) to provide pipeline route, well location, capture zone and ZOC overlays to 1:25.000-scale topographic maps. The maps of 50-day, 300-day, 400-day and 4000-day capture zones and results of ZOC interception of the pipeline are included as example maps of this report. A spill response procedure is developed for all groundwater operational zones. The overarching response strategy is to prevent spilled oil from entering the aquifer. The procedure continues until the threat to the aquifer is removed, and refers to information regarding site geology and aquifer characteristics included in this report. This analysis is undertaken to fulfill the requirements of the Environmental Management and Monitoring Plans defined in the Project’s ESIA in general. The development of the Groundwater Protection Strategy requires a systematic technical approach to the definition of groundwater sensitivity. These results will be used in the oil spill response planning. The following technical approach provides a summary of the core steps to be taken in realization of the groundwater protection strategy: Step 1. The major individual well field production zones

40

Huseyin Murat Cekirge:

Determination of Risk to Groundwater Aquifers from Crude Oil Pipelines

within the groundwater operational zones will be assessed in terms of their proximity and likely connectivity to the pipeline. Production zones within close proximity and / or in hydraulic connection with the pipeline will be selected for review and a groundwater database will be updated. Step 2. Based on the information provided in the literature reviews, the aquifer sensitivity will be assessed. This is likely to be guided by the importance and quality of the resource. Numerical risk values will be derived for the aquifer protection zones defined under Step 3. Step 3. Based on the complexity of the flow field, a suitable analytical or semi-analytical model will be chosen to delineate the 50 and 400-day capture zones and the Zone of Contribution (ZOC). USEPA Wellhead Protection Zone methodology can be used to delineate the protection zones, Figure 1. Step 4. The flow field and capture zones results will then be digitized and geo-referenced in the GIS database. The GIS will then be used to compute the level of sensitivity for each one kilometer section of pipeline. The results will be used to update the existing pipeline ERA (Environmental Risk Analysis) and will provide valuable information with regard to Oil Spill Response Planning; the procedures used in this evaluation are summarized in Figure 2. Figure 2. Procedures used in this paper.

2. Analysis 2.1. Preparation of the Model The WhAEM model requires a digital map base with coordinates. This was prepared in AutoCAD using the project’s pipeline route in UTM coordinates. Lines of equal hydrological potential were derived from previous reports and public records on each aquifer and added to the file. The map was exported from AutoCAD and imported directly into WhAEM in the base map were then traced and given the appropriate characteristic. The WhAEM model utilizes a series of entry screens to set model parameters for each model run. The necessary data: Aquifer Properties; Settings for Contouring; Settings for Tracing; and Well Properties. Well location data were input into a Geographic Information System (GIS). WhAEM model results having an interception of the pipeline with a capture zone or ZOC were input into the GIS by geo-rectifying the output image of the WhAEM model. All capture zone were traced into a GIS layer as a polygon for additional risk analysis and for incorporation into the Project’s GIS. Figure 1. Illustration of the zone of contribution; ‘Area of Contributing Recharge’; can be disconnected to the well site, [6] and [7].

2.2. Determining the Zone of Contribution (ZOC) The Zone of Contribution (ZOC) around a well is a surface representation of the area that contributes water to the well, and infers that any contamination within the zone of

International Journal of Environmental Monitoring and Analysis 2015; 3(6-1): 39-46

contribution would eventually reach the well. The ZOC is influenced by the three dimensional structure of the aquifer, in particular, by areas where there is confined or artesian flow and by areas of defined recharge. In other words, the ZOC will not be present if there is no connection to surface input or recharge. Therefore, the Zone of Contribution was determined as follows: 1. WhAEM was run for 4,000-days to indicate the capture zone for this time period. WhAEM is a steady state model so this indicates the general source of water for the well site but does not reflect the three-dimensionality of the aquifer, and, in particular, surface recharge areas; 2. Recharge areas for each aquifer were entered into the Geographic Information System (GIS); 3. The GIS was used to overlay and intersect the 4000-day ZOC with the aquifer’s recharge area; 4. Maps were created where the plume has the potential of extending further into the recharge area; this was estimated by visually extending the 4000-day plume. The sensitivity analysis in next section indicates that longer capture duration primarily alters the length of the capture zone, but has little change on its width. So the ZOC is estimated by extending the length of the 4000-day capture zone across the entire recharge area, but only slightly increasing its width as it is extended. The extension of the 4000-day aquifer (essentially the >4000-day capture zone) across the recharge area was mapped as a separate GIS layer; and 5. Maps of each aquifer were produced to indicate the ZOC based on the overlay of the 4000-day and the extended 4000-day capture zones. 2.3. Sensitivity Analysis WhAEM generalizes several aquifer characteristics that influence the transport of water through the aquifer. The assumptions are: Assumption of single uniform flow: The model assumes that there is no multi-layer flow. For instance, if wells are obtaining water from distinctly screened zones along the well pipe. All water is obtained across the length of a screened pipe indicating that a single uniform flow is a valid approximation. Assumption of a homogeneous aquifer: The WhAEM model considers that small inclusions of clay or gravel are inconsequential except if their size is on the order of the capture zone width or larger. This is not the case for the aquifers analyzed. The 50 - 100 m of screened water pipe crosses many smaller layers of various sediment types. Water is gathered across these layers in each well. Assumption of non-stratified aquifer: An aquifer having significantly different hydraulic conductivities for different parts of the aquifer along a single pipe length will yield different capture times. This information is not available for this study, but since a

41

single hydraulic conductivity is determined for each well, based on the transmissibility for the entire well sequence and the thickness of the aquifer (crossing many possible aquifer lenses), the assumption of non-stratification is supported. Assumption of non-transient (constant) flow rates: Summer pumping, as in agricultural areas, differs from winter pumping. For this study, the pumping rate is assumed to be constant year round. This results in a ‘conservative’ approach where the capture zone is maximized. To test the potential impact of changing WhAEM model entry parameters, an analysis was performed to determine the sensitivity of each value. The changes selected for analysis reflect large; 100% or more; to the base configuration values to ensure that changes can be visually observed. Lesser, more realistic changes in values, e.g. 10-20 %, would therefore produce relatively smaller alteration of the capture zone. 2.4. Environmental Risk Analysis This section first develops groundwater risk criteria and then applies the criteria to evaluate groundwater risk for each kilometer of pipeline passing through every aquifer. As appropriate, the existing Quantified Risk Assessment (QRA), which was made at the initial stage of the planning, is updated to reflect the knowledge gained during the analysis of each groundwater operational zone. The QRA was performed across the entire pipeline to provide guidance for environmental protection. In the case of groundwater, the QRA delineated the boundaries of each of the major aquifers to then provide a higher level of protection than in non-aquifer areas. The purpose of formulating this risk analysis is to determine if there are exceedences to the QRA values due to capture zones that intersect the pipeline and to provide further information to the Emergency Response Team in the event of an oil spill. 2.4.1. Risk Criteria There are several methods to determine the potential risk of aquifer contamination. The paper utilizes two evaluation procedures to each aquifer. These are: The Quantified Risk Assessment (QRA), [1] and The Groundwater occurrence-Overall lithology-Depth to groundwater (GOD) process, described in [4]. 2.4.2. QRA Risk Criteria Groundwater risk in the QRA report is based on potential pollution contact and the receptor that is in jeopardy of contact. In Table 1, this report presents the QRA criteria by including information about the capture zone and well site characteristics from this report. To determine the groundwater risk value for each pipeline kilometer, the Receptor and Contact values from Table 1 are cross-referenced using the matrix in Table 2.

42

Huseyin Murat Cekirge:

Determination of Risk to Groundwater Aquifers from Crude Oil Pipelines

Table 1. Criteria to determine valuation of risk to groundwater resources, quantified risk assessment (QRA), [1]. Rating

Criteria Receptor Evaluation 1.0 (Very High) Water supply to city or collection of small communities, to industrial complex, or to industrial agriculture. 0.7 (High)

Contact Potential Very high contact potential: Pipeline located within 50-day capture zone within an unconfined (non artesian) aquifer and depth to water level