Page 33: of Offshore Engineer Magazine (Dec/Jan 2013)

32 34

physical, environmental and ice regime Program reports that are available to the the 4.9m (16ft) water level for consistency

EPIC characteristics. public. with the original design analyses. The ‘All

A pipeline is not affected by ice Surveys use sidescan for ice gouge

Data Sets’ collection included gouges of gouges that do not cross it and this must detection, with a range of 50m on either both known and unknown age whereas the be addressed when defning the design side of the survey vessel trackline (100m two collections of Northstar-specifc data gouge depth. Additional factors can swath width). Multibeam sonar is then included only new (or known age) gouges.

infuence a pipeline’s response to seabed The qualitative goodness-of-ft of inves- used to map identifed ice gouges with ice gouging including pipeline operating tigated probability distribution functions effective swath widths ranging from 7.6 conditions; seabed soil characteristics; (PDFs) was assessed by comparing them to 42.7m, depending on the water depth gouge width; gouge orientation with against both the ice gouge depth data and returned acoustic beam signal. respect to the pipeline; trench backfll distribution histogram and the ice gouge The surveys in 2007 and 2011 showed conditions; and detailed interactions depth empirical cumulative distributions very signifcant seabed gouging during between the pipe, soil and ice keel. (see Fig. 2 and 3). Where qualitative the preceding 12-month period. INTEC- assessment could not distinguish the best SEA reported that an intense storm dur-

Studying the Northstar pipeline system ft, fnal selection of the PDF was based ing October 2006 produced high winds,

The BP Exploration (Alaska) Northstar on an Anderson-Darling (A-D) test. waves and a negative storm surge, while offshore Arctic pipeline system was The maximum ice gouge incision multiyear ice foes were present near the installed as a bundle and includes a depths observed along the route center- pipeline route.

nominal pipe size (NPS) 10-in. oil export line plus two, or more, offset survey lines The deepest observed 2007 ice gouge pipeline and a NPS 10-in. line supply- are recorded annually in the Northstar was 1.55m (5.1ft), exceeding the 100-year ing gas to the feld for reservoir pressure Development Pipeline Route Monitoring design ice gouge depth by a factor of maintenance. These pipelines extend 46%. However, the 1.55m (5.1ft) deep ice 18

CFC 1995-2011 - WD<4.9m outside the Alaskan coastal barrier gouge observed in 2007 was located 55m 16

Lognormal 14

Exponential islands and are exposed to more signif- (180ft) east of the pipeline centerline. 12 cant seabed ice gouge conditions than The maximum gouge depth observed 10 8 two other subsea Arctic pipeline systems directly above the pipeline that year was 6

Frequency operating in the Beaufort Sea. only 0.24m (0.8ft).

4

Two primary load conditions con- The 2011 survey reported 130 new 2 0 0 0.1 0.6 0.2 0.4 0.50.6 0.7 0.8 trolled design and trenching require- gouges – more than double the previ-

Water depth (m) ments: ice/seabed interaction in the ous highest record of 54 in 2002 – but deeper water zone outside the coastal no gouges exceeded the 100-year design

CFC 1995-2011 - WD<4.9m 100 barrier islands, and seabed permafrost ice gouge depth. The deepest ice gouge 90

Lognormal 80 thaw subsidence in the shallow lagoons. observed in 2011 was located near the

Exponential 70

Seabed ice gouging was observed from pipeline centerline, but not directly over 60 50 1995 through 1999, during each of the top of the pipeline and did not reduce 40

Percent yearly pipeline route bathymetry summer the minimum trench backfll soil thick- 30 surveys carried out before construction. ness above the pipeline. 20 10

Since installation, there have been 12 0 0 0.7 0.8 0.4 0.3 0.60.5 0.20.1 0.9

Going forward yearly surveys conducted. At the time

Water depth (m) the BP Northstar design was completed INTECSEA’s research shows that having

Fig. 2 Histogram and empirical cumulative in 1998, there were a total of 8 years of confdence in statistical models from distribution seabed ice gouge survey data in the vicin- multiple years of regional survey evalua- 80 ity of the pipeline route. tion can reduce conservatism in pre-

CFC 1995-2011 - WD>4.9m 70

Limit state design procedures for dicted extreme gouge-depths, potentially

Lognormal 60

Exponential pipe bending were used to calculate the resulting in reduced trenching and burial 50 minimum pipeline depth of cover to resist costs.

40 30 ice keel loads. A 2.13m (7ft) minimum INTECSEA concluded that since many 20 depth of cover was calculated based on years of site-specifc seabed ice gouge data 10 a 100-year return period maximum ice will not be available for new pipeline proj- 0 0 0.30.60.9 1.21.5 keel gouge depth of 1.07m (3.5ft) using ects, a balanced approach towards defning

Gouge depth (m) an exponential gouge depth distribution. reasonably conservative (deeper than

CFC 1995-2011 - WD>4.9m

The minimum depth of cover ranged from expected) design gouge depths and other 100 90 1.83-2.74m (6-9 ft), depending on location. design parameters affecting the pipeline

Lognormal 80

Exponential

INTECSEA studied three sets of data – safety is warranted. 70 60 the annual pipeline route monitoring sur- 50

Jonathan Caines is a pipeline engineering veys performed by Coastal Frontiers Corpo- 40

Percent Frequency 30 specialist and manager of small projects ration (CFC) along the route between 2000 20 with INTECSEA Canada in St. John’s, and 2011, the entire set of CFC surveys 10 0 0 0.30.60.9 1.21.5 conducted along the route between 1995

Newfoundland. Caines holds a B.Eng. in

Gouge depth (m) and 2011, and the full set of all available

Mechanical Engineering and an M.Eng. ice gouge data from all surveys in the area. in Oil and Gas Engineering, both from

Fig. 3 Histogram and empirical cumulative distribution

These data sets were further divided at

Memorial University of Newfoundland.

oedigital.com December 2013 | OE 35 034_OE1213_EPIC_Intecsea.indd 35 11/22/13 8:09 PM