Threat – Geohazard (Weather-Related Outside Force)
Outside force in the model includes ground/earth movement, flooding, cold weather and lightning. When buried pipelines cross-sloping terrain, there is a danger that soil movement (down the slope) will impose an excessive load on the pipe that could lead to failure. Slope or earth movements can be described as movement of earth materials (rock, debris, soil, etc) down-slope as a result of the pull of gravity. Defined as movement of mass of rock, debris, or earth down a slope, slope movement can be classified as falls, topples, slides, lateral, and flows. Complex landslides involve combination of two or more types of movement starting with material moving that eventually with combination of water (from rain or other water sources) form a heavier flow that may have more rocks, soils and other debris. The rate and severity of slope movement depends on the steepness of the slope, the presence of water, the stability of the soil and other loading or overburden applied to the slope due to construction and/or maintenance activities. Velocity within flowing mass decreases with depth and laterally. With combination of gravity pull, soil and/or rocks fall, slide, and/or flow along the surfaces and may endanger life safety and causing property damages.
Earthquake also causes earth movement. Earthquake damages/effect are categorize as tectonic and non-tectonic. Tectonic affects deals with fault / surface ruptures occur only during large earthquake. IT can cause severe damage to buildings, bridges, tunnels, canals and underground utilities. Non-tectonic are directly related to earthquake shaking and cause liquefaction, earthquake induced slope failures, landslides, tsunamis and seiches. Liquefaction typically occurs in soil with high groundwater table and usually, it is commonly observed in low areas or near water stream. Earthquake induced slop movement is often divided into falls (free-falling of rocks, debris, etc) and slides (movement along the surfaces).
Climate change can cause damages to the pipeline through heavy rain causing flooding, cold weather causing frost heaving, strong wind damaging pipeline segments supported on a suspension bridge and lightning strike on above ground pipeline segment or facilities. Flooding as a result of heavy rain and poor drainage; not only will add to the effect of slope movement but it also affects soil permeability. Cold weather freezes the ground. Water cannot drain away through still frozen ground and it accumulates in the soil. Frost heave occurs when the freezing of the soil results in the formation of layers of segregated ice at shallow depths. Thermal loading (expansion by the freezing of water in pores and fractures in cold regions), increases external load onto the pipeline. Strong wind from cold weather can also cause damage on above ground pipeline that are supported by suspension bridges. Lightning strikes on the pipeline can cause serious failure on the pipeline. Erosions are effect of moving rivers streams. It cannot be stopped however, the effect can be minimized. On some instances, failures can be mitigated by having good drainage for water to flow through, and provide insulation for pipeline segment that are susceptible to frost heave. Gas and liquid has the same likelihood model.
A suspended submarine pipeline is susceptible to dynamic excitation due
to vortex shedding of strong currents. If the frequency of vortex shedding is near the natural frequency of vibration of the suspended pipe segment, resonance may result in a damaging self-excited oscillatory motion of the line segment. In order to avoid this resonant response, the pipeline freespan length is restrained so that the natural frequency of the pipeline span will be out of the frequency range of vortex shedding.
The permissible length of the pipe suspended span is calculated based on
the natural frequency criterion as well as the allowable pipe stress
criterion.
Assuming that the pipe span segment is a simply-supported beam, the natural frequency (fn) of the suspended pipeline can be estimated using EI as the flexural stiffness of the pipe section, m as the mass of the
pipe section, and L as the length of pipeline span.
In order to safely avoid resonant response, it is general practice to set the
natural frequency of the suspended pipeline segment to be no less than
some multiple of the frequency of vortex shedding. Results of this vortex
shedding analysis indicate that the safe length of pipeline span.
The flexural strength of a pipeline free-span is evaluated by assuming that
the pipe span segment is a simply-supported beam. The permissible pipe
span length can be determined using M as the bending moment based on the allowable flexural stress and the pipe characteristics, and the submerged weight of the pipe.
The permissible length of pipeline free-span is determined based on the above variables.
US Natural Disaster Study
In the US, maps are available showing relative threats to pipelines from some common geohazards. While sometimes expressed with a only relative scale, the derivation of the rankings provides a way to generate frequency estimates for many of these phenomena, at least on a coarse—large geographical areas, possibly missing smaller but important features—level. It is useful to examine the methodology of establishing these hazard indices.
Excerpts from this ref [1018] are shown below to assist the risk assessor in determining the usefulness of such relative-scale information into a contemplated assessment. Note that, in the absence of more definitive information, a relative scale itself can be ‘grounded’ with frequency values and thereby used in preliminary exposure estimates (for example, an index value of 70 could be set to, say, 0.1 events/year, etc).
Then, the fraction of these events that could potentially damage an unprotected, structurally-weak component can be estimated. This exposure estimate is then used along with mitigation and resistance estimates to arrive at a failure frequency.
So, despite the ‘semi-quantitative‘ nature of this published information, it can serve as a good starting point for QRA. As of this writing, US databases area available [1018] showing hazard indexes for:
- earthquake (HER = Earthquake Hazard Rank)
- hurricane (HHR = Hurricane Hazard Rank)
- tornado/storm (TSRR = Tornado/Storm Hazard Rank)
- flood/scour (FHR = Flood Hazard Ranking)
- landslide (LHR = Landslide Hazard Ranking)
- other (lightning and snow depth; OHR = Other Hazard Risk)
National Pipeline Risk Data
| Hazard | Variables Included | Methodology | Notes |
| Hurricane | Historical count | 94 year history of hurricanes per coastal county | 2 |
| TSRR | Historical count | Number of occurances over 30 years per one degree box | 3 |
| Landslide | swelling clays, landslide incidence, susceptibility, subsidence | LSHR = 0.3 (clay) + 0.4 (incidence) + 0.2 (susceptibility) + 0.1 (subsidence) | 6 |
| Earthquake | Spectral response acceleration coefficient | Based on single, complex variable ranked 0-100 | 4 |
| Other | 30 year Mean annual lightning strike density; Snow depth with 95% chance of not being exceeded | OHR = .5 (lightning strike) + 5 (snow depth) | 5 |
| Flood | Annual flooding frequency, potential scour depth | FHR = 0.5(flooding) + 0.5(scour depth) | 1 |
Table Notes
- For the Annual flooding frequency layer one-kilometer grid cells were assigned the following values based on the annual chance of flooding:
Frequent (5O-100%): Flooding = 100
Rare (O-5%): Flooding = 33
Occasional (5-50%): Flooding = 67
No Flooding: Flooding = 0
These values were then multiplied by the percentage of area they covered for each soil map unit. The percentage values were summed to give the value for each soil map unit. A grid of these values was created and then ranked from 0 to 100. For the Potential scour depth layer one-kilometer grid cells were ranked based on their value (potential scour depth in feet).
Highest value Scour depth = 100 Lowest Value Scour depth = 0 - The total number of direct and indirect landfalling hurricanes per coastal county was used from 1990 (assume typo—should probably be “1990”) until 1994. From the county baaed polygon coverage, a point coverage was derived. From this point coverage first a Triangulated Irregular Network (TIN) and then a continuous surface grid was created, in order to more appropriately represent the hazard without the use of political boundaries. These numbers were ranked from zero to 100, where 100 represents the highest number of land-falling hurricanes and zero represents the lowest number of land-falling hurricanes.
- The centroids of the one-degree cell areas were used to generate a Triangulated Irregular Network (TIN). This resulted in a continuous surface that more naturally depicts the distribution of tornado events. A grid was created at a resolution of one kilometer from the TIN. The values were ranked from zero to 100, where 100 represents the highest number of tornadoes and zero represents the lowest number of tornadoes.
- The spectral response acceleration coefficient is an indicator of the probability of receiving specific intensities of ground shaking from earthquakes. For the EHR the spectral response acceleration coefficient at a period of 0.3 seconds expressed as a fraction of gravity with a 90% chance of not being exceeded in 50 years is used. The data are prepared by the U.S. Geological Survey (USGS) for the NEHRP Recommended Provisions for the Development of Seismic Regulations for new Buildings.
- Lightning strike density is expressed in the mean annual number of flashes per square kilometer. Contour lines were digitized from a very small scale map. The areas in between the contour lines were given the mid-value of the class. These values are ranked from zero 100 for the country, where 100 represents the highest lightning strike density and zero represents the lowest lightning strike density. The 95% annual nonexceedence probability was calculated for 239 weather station in the United States. From this point coverage first a Triangulated Irregular Network (TIN) and then a continuous surface grid was created in order to more appropriately represent reality. These numbers were ranked from zero to 100, where 100 represents the highest snow depth and zero represents the lowest snow depth.
- The LSHR values were ranked from zero to 100, where 100 represents the highest ground failure hazard and zero represents the lowest ground failure hazard.
While relative indices like these are not ready for direct inclusion into a modern risk assessment, the underlying methodology provides insights into the phenomenon and current abilities to forecast them.
Test of Time Evidence
Geohaz essay for ASME “Invited Perspectives”
Introduction
Earth movements are one of multiple threats to a pipeline system’s integrity (see Fig 1). The term ‘geohazards’ is often used to encompass all types of potentially damaging earth movements, from landslides, erosion, subsidence, and seismic activity to buoyancy, scour, and seabed instability. Each of these phenomena warrants treatment in a pipeline risk assessment for a pipeline system.
Several key players have been involved in managing pipeline geohazard risks. The geotechnical scientist/engineer has been perfecting his craft; getting better and better at understanding and forecasting geohazard events that could damage pipeline components. The pipeline designers continue to improve structural engineering models to better design components to withstand new forces. Risk assessment specialists have improved the way in which risk can be quantified while spill/release specialists have continued to refine their models that detail consequence potential
However, until recently, there has been a bit of a gap in putting all the pieces together from these varied specialists into a single, comprehensive framework to truly understand pipeline geohazard risk. This essay reports on the key breakthrough that now allows the work of these disciplines to combine into a powerful risk assessment.
GeoHazards
There are many good books written on geohazard risk. Nonetheless, even the well-tested methods discussed in current texts can benefit from a recent breakthrough in quantifying risk. That breakthrough is the realization that, just as in the design phase of engineered systems, a risk assessment has to independently measure three things in order to gain a complete understanding of PoF.
The comparison to the design process goes like this: we know that “Mother Nature hates things she didn’t create”. So, its best to acknowledge that our new, engineered installation will be under constant attack (corrosion, land movements, outside forces, fatigue, etc). There are two basic ways we can respond to this attack and prevent failure. We can defend against the attacks or we can make our system so strong that it can absorb the the damage from the attacks without failing. For practical reasons, designs typically use both.
The interplay between these three elements—what is attacking; how effective are the defenses; what happens if the attack reaches the component—is the key to measuring PoF. We must understand the contribution from each in order to really understand PoF.
Armed with good PoF estimates, we can couple them with CoF estimates and arrive at much more meaningful and useful risk estimates. Each location along a pipeline can have a $/year Expected Loss (EL) value assigned from each potential threat. Imagine the improvements in risk management that are possible once risks are fully understood and quantified in this way.
Measuring PoF
All plausible failure mechanisms must be included in the assessment of PoF. Each failure mechanism must have each of the following three aspects measured or estimated in verifiable and commonly used measurement units:
- Exposure (attack)—the type and unmitigated aggressiveness of every force or process that may precipitate failure
- Mitigation (defense)—the type and effectiveness of every mitigation measure designed to block or reduce an exposure
- Resistance—a measure or estimate of the ability to absorb damage without failure, once damage is occuring
Measuring exposure independently generates knowledge of the ‘area of opportunity’ or the aggressiveness of the attacking mechanism. Then, the separate estimate of mitigation effectiveness shows how much of that exposure should be prevented from reaching the component being assessed. Finally, the resistance estimate shows how often the component will failure, if contact with the exposure occurs. In risk management, where decision-makers contemplate possible additional mitigation measures, additional resistance, or even a re-location of the component (often the only way to change the exposure), this knowledge of the three key factors will be critical.
Units of measurement are transparent and intuitive. In one common application of the exposure, mitigation, resistance triad, units are as follows. Each exposure is measured in units of ‘events per time and distance’, ie events/mile-year, events/km-year, etc.
An exposure event is an occurrence that, in the absence of mitigation and resistance, will result in a failure. To estimate exposure, we envision the component completely unprotected and highly vulnerable to failure (think ‘tin can’ wall thickness). So, almost any kind of earth movement involving a pipeline is an event.
Mitigation and Resistance are each measured in units of % representing ‘fraction of damage or failure scenarios avoided’. A mitigation effectiveness of 90% means that 9 out of the next 10 exposures will not result in damage. Resistance of 60% means that 40% of the next damage scenarios will result in failure, 60% will not.
For assessing PoF from time-independent failure mechanisms—those that appear random and do not worsen over time—the top level equation can be as simple as:
PoF = exposure x (1 – mitigation) x (1 – resistance)
With the above example units of measurement, PoF values emerge in intuitive and common units of ‘events per time and distance’, ie events/mile-year, events/km-year, etc.
As an example of applying this to failure potential from landslides, let’s assume that the following inputs are identified for a hypothetical pipeline segment:
- Exposure (unmitigated) is estimated to be 0.2 rainfall-initiated landslide events per mile-year, ie, an event every 5 years.
- Using a mitigation effectiveness analysis, SME’s estimate that 1 in 10 of these exposures (attacks) will not be successfully kept away from the pipeline by the existing mitigation measures. In other words, an overall mitigation effectiveness of 90%.
- Of the exposures that reach the pipe, despite mitigations, SME’s perform an analysis to estimate that 1 in 4 will result in failure, not just damage. This estimate includes the possible presence of weaknesses due to threat interaction and/or manufacturing and construction issues. So, the pipeline in this area is judged to be 75% resistive to failure from this these landslide events, once contact occurs.
These inputs combine for a simple PoF[1] calculation:
(0.2 landslide events per mile-year) x (1 – 90% mitigated) x (1 – 75% resisted) = 0.005 failures per mile-year from landslides
This suggests a landslide-related failure about every 200 years
This is a very important estimate. It provides context for decision-makers. When subsequently coupled with consequence potential, it paints a valuable picture of this aspect of risk.
Note that a useful intermediate calculation, probability of damage (but not failure), also emerges from this assessment:
(0.2 landslide events per mile-year) x (1 – 90% mitigated) = 0.02 damage events/mile-year
This suggests landslide-related damage occurring about once every 50 years.
This damage estimate can be verified by future inspections such as in-line inspection (ILI). Differences between the actual and the estimate can be explored: eg, if the estimate was too high, was the exposure overestimated, mitigation underestimated, or both? This is a valuable learning opportunity.
Geohazard Exposures
An interesting aspect of geohazard as an exposure–an ‘attack on’–pipeline integrity, is the distinction between measuring the geohazard event vs measuring its effect on a pipeline. For example, the initiating geohazard event could be ‘flood’, resulting in subsequent events of
- scour,
- bank erosion,
- avulsion,
- debris transport,
- and others.
We often have published recurrence intervals for these initiating events and/or the secondary events. These published frequencies are the first step in our estimates of the PoF attack frequency. However, distinct from the initiating geohazard event, the events for which we really seek an estimate of attack frequency are the pipeline’s integrity threatening events. The pipeline-threatening events to be related to the above list of geohazard events are
- lack of support (increasing the gravity loading)
- buoyancy (creating an uplift force)
- lateral loadings (from flowing current and debris)
- oscillations (fatigue loadings)
These will be some fraction of the geohazard event frequencies since not every geohazard event will generate sufficient forces to threaten the pipeline.
Geohazard Mitigation
Much geohazard risk reduction occurs in the design phase. Realtime mitigation is often not a prime method to reduce PoF but nonetheless an available option at times. Stress relieving, rockslide barriers, dewatering, bank stabilization measures, concrete coating, anchoring systems are examples of mitigation against geohazard threats. While some might argue that these, rather than being barriers–blocking the exposure—some of these are actually modifying the exposure frequency. Bank stabilization, for instance, is not actually a barrier but rather a pre-emptive action. Nonetheless, in the interest of transparency, it seems more useful to include as a mitigation anything that either blocks or reduces an exposure.
Geohazard Resistance
The ability of a structure (pipeline component, in this case) to resist an external force of any kind is fairly well understood, although it is not a trivial estimation process. The estimation is further complicated by the potential presence of defects in the structure, reducing its ability to resist a load. Examples include dents, gouges, corrosion, ovalities, cracks, and others. We can employ a range of analyses rigor, from simple informed-estimation to finite element analyses using sophisticated calculation routines.
For initial stage risk assessments, we should at least use methods that reasonably approximate the actual ability of the structure to absorb damages in light of all that is known or reasonably inferred about the component. This can be done in a broadcast-fashion, where a few strength-approximating routines are efficiently applied to hundreds of miles of pipeline via database algorithms.
Geohazard CoF
A more detailed risk assessment could pair each failure mechanism with a corresponding set of consequence scenarios. A simpler assessment will often use a single CoF estimation routine to link to all possible failure mechanisms, as illustrated in Fig 1.
A challenge in pipeline geohazard risk analyses is the need for primary, secondary, etc, consequence evaluation. As previously noted, the initiating geohazard event may not generate the most consequences. It is often the potential subsequent events that hold the majority of consequent potential.
For instance, the risk of dam failure does not always include a quantification of all related damages that may unfold for days, weeks, or longer after the actual failure. While the geohazard expert can determine the height, flowrate, an duration of released flood waters and the ensuing channel damages, he is rarely the expert on population escape potential, property values, clean up costs, long term environmental and societal ramifications, etc, all of which are legitimately a part of the dam failure.
Similarly, ignition potential from a pipeline release, hazard zone generation, thermal radiation levels, service interruption costs, and many others, are not typically in the geohazard expert’s realm of expertise.
Fortunately, the risk assessment framework now available to practitioners provides clear placeholders for each discipline’s experts to provide input. That collective of expertise makes the entire risk assessment more accurate and defensible.
Geohazard EL
Finally, the modern risk assessment framework provides for the monetization of risks—a measurement unit with a powerful common denominator.
Risk expressed in this fashion is called “expected loss” (EL). It encompasses the classical definition of risk: probability x consequences, but expresses risk as a probability of various potential consequences over time.
An EL analysis captures the high-consequence-extremely-improbable scenarios; the low-consequence-higher-probability scenarios, and all variations between. It does this without overstating the influence of either end of the range of possibilities, resulting in a fair representation of risk and providing powerful risk management decision-support.
Key Take-Aways
The improvements in risk assessment—including the ability to efficiently quantify geohazard PoF—have opened new doors in managing risks. Armed with a modern risk assessment, risk management becomes much more transparent and even exciting. Seeing actual risk levels and generating cost/benefit analyses of proposed risk reduction actions provides clarity and, often, some ‘ah-ha!’ moments.
References
Muhlbauer, W.K. (2015). Pipeline Risk Assessment: The Definitive Approach and Its Role in Risk Management, Expert Publishing, LLC, 580p
[1] More correctly, a FoF, frequency of failure