This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Consequence modelling for effective pipeline emergency response planning

Author : Anthony Salazar, BakerRisk

14 January 2022

A comprehensive emergency response plan (ERP) is a crucial component of pipeline safety management for protecting the public and property surrounding pipelines that contain hazardous materials. But what constitutes a comprehensive ERP for pipeline emergencies?

Figure 1 – A fish-mouth rupture which led to a full-bore equivalent pipeline release
Figure 1 – A fish-mouth rupture which led to a full-bore equivalent pipeline release

(Click here to view article in digital edition)

Typically, ERP guidance must remain relatively generic since pipeline incidents can vary significantly based on a pipeline’s operating conditions and the conditions surrounding the pipeline such as topography, buildings, residences, roads, railroads, etc. So how can ERPs be enhanced to ensure they are comprehensive, and the public is sufficiently protected from potential hazards?

One such way ERPs can be enhanced is through an improved understanding of the consequences from pipeline loss-of-containment events for specific pipelines of interest. Consequences can be represented by distances to thresholds of concern such as flammable concentration, toxic concentration, explosion overpressure, or thermal radiation endpoints of interest.

Pairing the consequence extents with geographic information system (GIS) data in the form of contours overlaid on a site aerial image provides a detailed visualisation of hazard thresholds along an entire pipeline system route. This visual improves the understanding of consequences for emergency responders and pipeline operators. This article discusses a methodology for creating these visuals, what they mean, and how they are useful to enhance emergency response planning. Additionally, the consequence analysis utilised in this methodology largely overlaps with the scope of performing a pipeline transportation risk analysis; therefore, if one is required, they can efficiently be done concurrently.

Consequence Analysis

In order to create these visuals, the first step is performing the consequence analysis itself. This can be done by addressing the following questions:

1)  What are the hazards?

2)  What needs to be modelled?

3)  How is it modelled and what do we gain from it?

What are the hazards?

One of the first steps in Pipeline Safety Management is understanding the potential hazards a release from a pipeline may pose. The primary interest of this article is loss-of-containment (LOC) events in which material is released from the pipeline. Releases can be liquid, vapour, or two-phase flow depending on the material, temperature, and operating pressure of the pipeline. Hazards from a LOC event can be broken up into four main types as shown in Table 1.

Release Scenario

Consequence

Hazard

Non-ignition

Vapour cloud travelling to its maximum extents

Exposure to toxic materials or potential asphyxiation

Immediate ignition

Jet/pool fire depending on phase of release

Exposure to sustained thermal loads

Delayed ignition without congestion and confinement1

Flash fire, followed by jet/pool fire until isolation and depressurisation

Exposure to short duration thermal loads throughout flammable extents of vapour cloud and sustained thermal loads to subsequent jet/pool fire

Delayed ignition with congestion and confinement1

Explosion, followed by jet/pool fire until isolation and depressurisation

Exposure to overpressure and sustained thermal loads to subsequent jet/pool fire

[1] Congestion and Confinement is defined by the Baker-Strehlow-Tang (BST) methodology, [1][2][3]a leading edge explosion model that has been validated and tested with large-scale vapour cloud explosion field tests.

This article focuses on flash fire and jet/pool fire hazards; however, similar methodologies can be used to assess toxic and explosion hazards as well.

Figure 2 – Release scenario combination tree with example path
Figure 2 – Release scenario combination tree with example path

What needs to be modelled?

The process conditions and physical attributes of the pipeline are required to understand and model these hazards. This includes the pipeline’s operating pressures, temperature, material composition, flow rate, and diameter. The pressure, temperature, and composition are required for modelling, while the flow rate and diameter provide limits of what needs to be modelled.

Some guidance requires response plans to account for worst-case discharge scenarios. However, since smaller hole sizes are more likely events, it can be helpful to model a range of hole sizes up to full-bore equivalent to represent a variety of failures that would have corresponding release rate.

If the release rate predicted for the range of hole sizes based on pressure, temperature, and composition exceeds the maximum possible flow rate (MPFR) through the pipeline segment, the release rate is refined to the MPFR for consequence calculations. The MPFR can be defined by flow limitations in the pipeline’s supply, such as upstream pump rates.  This forms the basis for the Maximum Credible Hazard (MCH) or Major Accident Hazard (MAH) scenario.

Historically, there have been full-bore LOC events. Figure 1 shows an example of one of the incidents BakerRisk has investigated over the past 35+ years. To prevent emergency response planning gaps arising from preparing for less than worst-case scenarios, operators should consider evaluating multiple release sizes up to the MPFR release scenario.

In addition to understanding what hole sizes to model, some assumptions need to be made around other parameters to perform discharge and dispersion calculations. These assumptions and decisions can vary depending on the specific analysis goals and needs of the stakeholders, but can generally be summarised as follows:

1.  Material Composition: Pipeline systems should be split into segments based on material composition as necessary. If multiple compositions are transported through a pipeline, the material with the highest consequence can be conservatively used for simplicity. However, a more rigorous analysis may review the consequences for each composition.

2.  Pressure:  An average pressure can be used to model a pipeline release. Alternatively, a low and high operating pressure can be modelled for further granularity. While this article will refer to the method of using a low and high pressure when modelling release scenarios, both options can be representative of release scenarios depending on pipeline operations.

3.  Above/Below Grade: Whether a pipeline is above or below grade may significantly affect the consequences of a release. Both above and below-grade scenarios should be modelled so that either one of the calculated consequences can be utilised. Above-grade releases are conservatively modelled as horizontal releases since this typically results in the farthest dispersion and thermal radiation distances.  Below-grade releases are modelled at a 45° angle from the ground that accounts for the geometry of the crater, which is formed from release expansion energy.

4.  Weather Conditions: The weather (wind speed and stability) can greatly affect the distance vapour clouds will travel. Local weather statistics can provide the most accurate results. To limit complexity, two weather conditions representing day and night may be used such as D and F Pasquill atmospheric stabilities with wind speeds of 3.0 m/s and 1.5 m/s, respectively.

Figure 3 – Example 3D view of vapour cloud
Figure 3 – Example 3D view of vapour cloud

5.  Release Duration: The duration of a release can be assumed to be 60 minutes to account for the time it takes to identify, locate, and isolate the LOC. However, it is important to note that depending on the release composition, conditions, and magnitude, full consequences may be achieved in seconds or minutes.

A visualisation of the different combinations of parameters can be seen in Figure 2.

How is it modelled and what do we gain from it?

Scenario modelling is performed using dispersion and blast modelling software to evaluate liquid spills, vapour cloud dispersions, explosions, jet/pool fires, and toxic releases. The dispersion and jet/pool fire models are based on industry available data and refined based on data collected from full-scale testing conducted by BakerRisk and published previously in the Process Safety Progress Journal [4]. The dispersion modelling and jet/pool fire modelling software utilised should be a validated and industry accepted consequence software capable of determining flammable concentration and thermal radiation extents.

Dispersion distances may significantly vary based on the surface and topography around pipelines where liquid released materials can run off and/or pool. If a pipeline travels along a non-flat terrain or is near many rivers and streams, it may be necessary to utilise terrain modelling in calculations.  However, if the area around a pipeline is relatively flat, a flat 2-D model is typically sufficient to calculate dispersion distances for highly volatile liquids. In general, this article refers to the method of utilising a flat model for dispersion modelling, but the same methodologies can be applied with dispersion results from utilising a terrain model.

The consequence results for each scenario combination, as described in Figure 2, include the following calculated data:

• Discharge Rate

• Flammable dispersion extents after steady state are reached

   - Provided as distance to flammable concentrations of interest, such as UFL (upper flammable limit), LFL (lower flammable limit), 0.5 LFL, and/or 0.1 LFL

• Thermal radiation extents after steady state is reached

   - Provided as distance to thermal fluxes of interest, such as 37.5, 12.5, and/or 4.0 kW/m2

• The size and concentration of the cloud evolving over time

Consequence Extent Contours

A streamlined method of conveying the potential area of impact is to show the extents in graphical form. This allows the data to be documented and readily accessible.  The results shown in this article case study are produced using BakerRisk’s [5] dispersion and blast modelling software SafeSite3G©Note that these results are an example only and do not represent an operating pipeline.

Dispersion calculations in SafeSite3G© form a 3D cloud (Figure 3), which can also be represented as 2D slices such as a top-down view or a side view.  These representations provide a detailed picture of the shape of the cloud at a single potential release location but can be streamlined even further. To better convey these extents along the length of a pipeline, the flammable and thermal extents are rotated 360° to account for potential wind directions. For a pipeline, this is represented along the entire length of pipeline, utilising the GIS data on the pipeline route, as shown in Figure 4.

These contours represent the maximum extent of a LOC event along the pipeline route – they do not represent a single event, or secondary impacts. While contours can be generated for different scenarios of interest, only maximum extent consequences are shown for this example. Maximum extents were defined as the MPFR release (i.e., MCH or MAH).

In this example, flammable contours are broken up into three extents of interest as described below:

• 100% LFL – Concentrations at or above this level (until the UFL) are flammable and will ignite if they find an ignition source.

• 50% LFL – The extent where localised flammable concentrations have the potential to be greater than the LFL due to specific scenario circumstances (such as topographically low areas on the surface where a heavy cloud may accumulate).

• 10% LFL – This is the typical sensitivity limit on personal gas monitors.

The same methodology can be applied to jet/pool fires where a release was immediately ignited or what could potentially follow a flash fire or explosion scenario. This produces thermal radiation extent contours which are analogous to those in Figure 4, but to different thresholds of interest. An example of a thermal threshold of interest could include the radiation required to cause severe burns or ignite wood after exposure.

Enhancing Emergency Response Planning

Figure 4 – Example flammable dispersion extent contours
Figure 4 – Example flammable dispersion extent contours

Looking at Figure 4, one can easily determine vulnerable areas around the pipeline with respect to flammable vapour. For the scenario defined and considering any potential release location and wind direction, the area within the red line is susceptible to reaching LFL concentrations, the orange line indicates 50% LFL concentrations, and the green line indicates 10% LFL concentrations. This is particularly useful from a land use planning perspective for three main reasons:

1) Avoiding sensitive areas when routing new pipelines, to the extent practical.

2) Ability to make informed decisions on purchase of land for stand-off distance purposes if an operationally viable option.

a. Post construction, contours are useful in understanding potential new high consequence areas that can arise due to urban development expansion.

3) Response personnel can look at the contours and quickly determine exclusion areas based on the worst-case consequences from a pipeline release without needing a detailed understanding of the pipeline operations.

Contours used for these purposes will need to be based on pipeline GIS data so that images can be created at a high resolution and utilised in specific locations throughout the entire pipeline.

As pipeline networks grow and adapt to accommodate a cleaner mix of energy, this methodology can aid in creating a safer sustainable future.

Conclusion

Consequence modelling of pipeline LOC events can be used to quantify consequences and in turn utilised for Emergency Response Planning. Specifically, consequences can be quantified for flash fire and jet/pool fire hazards with distances to flammable concentration and thermal flux extents of interest. Once consequences have been calculated for a variety of release scenarios utilising dispersion and jet fire modelling software, the worst-case scenario extents can be utilised to create contours around the pipeline to extents of interest.

Pairing the consequence extents with geographic information system (GIS) data in the form of contours provides a detailed visualisation of hazard thresholds along an entire pipeline system route. These high-resolution contours can aid in land use planning and emergency response plans by providing pipeline owner/operators and emergency response personnel with a better understanding of what pipeline hazards exist and more importantly, precisely where they exist, so public safety can be maximised. 

References 

[1] Baker, Q.A., M.J. Tang, E.A. Scheier and G.J. Silva (1996) “Vapor Cloud Explosion Analysis,” Process Safety Progress, 15(2): 106-109.

[2] Baker, Q.A., C.M. Doolittle, G.A. Fitzgerald and M.J. Tang (1998) “Recent Developments in the Baker-Strehlow VCE Analysis Methodology,” Process Safety Progress, 17(4): 297-301.

[3] Pierorazio, A.J., J.K. Thomas, Q.A. Baker and D.E. Ketchum (2005) “An Update to the Baker-Strehlow-Tang Vapor Cloud Explosion Prediction Methodology Flame Speed Table,” Process Safety Progress, 24(1): 59-65.

[4] Zhao J, Rowley J. “Liquid jet fire: The impact of rainout on the predicted hazard.” Proc Safety Prog. 40(3) September 2021; 98-104. https://doi.org/10.1002/prs.12237

[5] SafeSite3G© Baker Engineering and Risk Consultants, Inc., San Antonio, TX. https://www.bakerrisk.com/products/software-tools/safesite/

[6] Fergusson, A.I., A.J. Salazar, D.W. McQuade, (2021) “Identifying and Utilizing Pipeline Consequences for Pipeline Safety Management,” presented at the GPA Midstream Association conference, September 26-29, 2021.  San Antonio, TX.

About the author:

Anthony Salazar is a Process Safety Consultant at BakerRisk where he has been performing pipeline risk analysis since he joined in 2020 after completing his studies in Chemical Engineering at the University of Illinois at Urbana-Champaign. His main technical practice involves consequence- and risk-based facility siting studies (FSS) with a focus on transportation hazards including major pipelines. To discuss pipeline risk assessments, Anthony can be contacted at ASalazar@bakerrisk.com.

Co-authors on this article were David W. McQuade, Targa Resources and Alexander Fergusson, Chicago Operations Manager – BakerRisk.

 Anthony Salazar, Process Safety Consultant - BakerRisk


Contact Details and Archive...

Print this page | E-mail this page