Page 352 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Case studies 14/327
I. About 38 percent of reportable leaks are of a size to pose a threat to From these input and the equations and assumptions shown
a recreational water supply. in Ref. [43], the calculations shown in Table 14.43 were
2. Ofthose leaks, -25 percent would contaminate the receptor. This made.
is determined by characterizing the various lengths of such recep-
tors present within each tier. Each length within each tier is
assigned a probability, indicating that length’s vulnerability. In Case Study E: Sour gas
aggregate, these compute to be the equivalent of about a 25 per-
cent probability all along the pipeline. The following is an excerpt from Ref. [22], a quantitative risk
3. The industry average leak rate applied to this pipeline predicts 35 analysis of anatural gas gathering system located in southwest-
leaks and, hence, about I3 (38 percent of35) would heofsufficient ern Wyoming. This describes an analysis approach and some
volume, and -2.8 would occur at the right location to contaminate assumptions used in assessing risks from a toxic gas system and
one of these receptors.
expressing those risks in absolute terms.
Further discussion of how this receptor is modeled can be found in The objective of the analysis was to determine the risk the
Attachment C ofthis report [Appendix F ofthis book]. pipeline and associated wells pose to the public population
along the pipeline route. This required the completion of four
Prime agricultural land contamination major tasks.
A spill size of 500 bbl over prime agricultural land is viewed as Task 1 : Determine potential pipeline and wellhead accidents
impacting agricultural lands, based on the potential for spread of a that could create life-threatening hazards to persons located
rapid release to impact 1/4 acre of agricultural lands. Further discus- near the pipeline or well sites.
sion of how this receptor is modeled can be found in Attachment C of Task 2: Derive the Frequency of occurrence (probability) of
this report [Appendix F ofthis book]. each accident identified in the first task.
Task 3: Determine the consequences of each accident identi-
Wetlands contamination fied in the first task.
Task 4: Combine the consequences and the probability of
A spill size of 500 bbl over wetlands is viewed as impacting the wet- occurrence of each accident to arrive at a measure of public
lands. This threshold is set as a level which would potentially over- risk created by the pipeline and well network.
come the natural processes of volatilization and adsorption, and cause
serious degradation of high quality impacts. Discussion of how this
receptor is modeled can be found in Attachment C of this report The natural gas being produced and transported through the network
[Appendix F ofthis book]. varies in composition from one section of pipeline to another, accord-
ing to the gas produced from each well. However, all pipeline sections
transport natural gas containing some hydrogen sulfide. The pipeline
Summav of results creates no hazards for persons near the pipeline or well sites as long as
the sour natural gas is contained within the pipeline.
Post-mitigation impact frequencies are calculated to be IO to 30 times
lower than pre-mitigation and industry average frequencies. The fre- Accidental releases of sour natural gas from the well: pipeline
quency reduction is not constant since different permutations of leak network could create potentially life-threatening hazards to per-
frequencies, spill size frequencies, and lengths-impacted are com- sons near the location of the release. Due to the presence of hydro-
bined. gen sulfide in the natural gas, the vapor cloud created by a release
of gas to the atmosphere would he toxic as well as flammable.
The following tables show the results of all frequency estimates for all Persons inhaling air containing toxic hydrogen sulfide vapor could
impacts. Case 4 in all tables shows the estimate for post-mitigation be fatally injured if the combination of hydrogen sulfide concentra-
results. Other cases are included for comparison. Table 3 shows overall tion and time of exposure exceeds the lethality threshold. If the
frequencies for all cases andTable 4 shows segment-specific frequencies cloud is ignited, persons in or very near the flammable vapor cloud
forall cases.Tables5 and6focusonCases 3 and4andpresentprobabili- could be fatally injured by the heat energy released by the fire.
ties (in slightlydifferent formats thanTables3 and 4) ofimpacts.
The frequency of occurrence of each potential pipeline accident
identified in Task 1 was estimated from historical pipeline failure rate
Case Study D: highly volatile liquids data gathered by the US. Department of Transportation. Event trees
were then used to estimate the percentage of releases of various sizes
This case study is the example presented in Ref. [43]. That ref- that would create a toxic or fire hazard. For example, it was estimated
that 50 percent of moderate-sized releases of sour natural gas from the
erence describes a recommended methodology for calcuiating pipeline do not ignite but do create a toxic cloud; IO percent ignite
hazard zones for highly volatile liquids (HVLs). This report immediately on release and create a torch fire; and 40 percent ignite
appears to have been produced for the National Energy Board after some delay, thus creating a toxic cloud followed by a torch fire.
ofCanada.
The example is a 25-km-long propane pipeline with The frequency of sour gas well blowouts was derived from sour gas
0.1589-m inner diameter, operated at 9928 kPa, in a popula- well historical data. The largest documented data base covers wells in
tion class 1 (rural, 5 dwelling units). The scenario is a full- the Province of Alberta, Canada. According to the data, an uncon-
rupture event with the following assumptions: a frequency of trolled sour gas well blowout would occur with a frequency of
3.55E-06 blowouts per well per year. This failure rate is for wells
failure of 2.0E-03, wind speed 4 m/s (probability of the popu- equipped with subsurface safety valves (SSSV’s). All the wells in the
lation being exposed based on wind direction), probability Wahsatch network will be equipped with SSSV’s.
of ignition = 12%, probability of exposure = 11% (taking
into account moving populations), and mortality rate = 0.5 Computerized consequence models were used to calculate the
fatalities per event. extent of potentially lethal hazard zones for toxic vapor clouds