Long Beach City Council
March 21, 2023

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

5. Subsea equipment
a. Author: Bjørn Søgård
b. Work verified by: Rolf Benjamin Johansen

6. Drilling and wells
a. Author: Leif Halvor Moen
b. Work verified by: Axel Stang Lund, Lars Tore Haug

7. Mooring system
a. Author: Bjørn E. Sogstad
b. Work verified by: Siril Okkenhaug

Images used are DNV property unless otherwise indicated.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
2 DEGRADATION MECHANISMS
Relevant degradation mechanisms are presented in Section 2.1.

All mechanisms, which are relevant for the five areas m entioned in Section 1, are included in this chapter.

2.1 Degradation mechanisms
A degradation m echanism is here defined as a disintegration of a metallic m aterial due to th e impact of the operating environm ent and forces. Degradation can be due to erosion, corrosion or stresses induced by cyclic/dynam ic loads an d other specific environm ental impacts. The degradation mechanism can result in metal loss (as uniform or localised attacks) or cracking (e.g. fatigue, stress corrosion cracking, embrittlement). Degradation mechanisms related to metal loss and fatigue are typically tim e dependent (e.g. wall thinning), whilst cracking m echanisms are of more abrupt nature.

2.1.1 Erosion
Erosion can be defined as physical rem oval of surface m aterial due to num erous individual impacts of solid particles, liquid droplet or im plosion of gas bobbles (c avitation). Erosion is a time dependent degradation mechanism, but can sometimes lead to very rapid failures.

In its mildest form, erosive wear app ears as a li ght polishing of the upstream surfaces, bends or other stream-deflecting structures. In its worst form, considerable material loss can be obtained.

2.1.2 Corrosion

2.1.2.1 General
Corrosion is caused by a chem ical (or electrochem ical) reaction between a m etal and its environment that produces a dete rioration of the m aterial a nd som etime its properties. For corrosion to occur, the following basic conditions must be fulfilled:
• metal surface exposed to enviro nment (bar e steel in physical contact with the environment)
• electrolyte (e.g. water containing ions, the electrolyte must be able to conduct current) • an oxidant (a chemical component causing corrosion (e.g. oxygen, carbon dioxide)

If one of these conditions is not present, no corrosion will occur.

Table 2-1 summ arises prospect ive corrosion mechanism s for subsea oil and gas production equipment.

The presence of organic acids and sulphur containing com pounds (e.g. elem ental sulphur) m ay aggravate the corrosion in the system.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Table 2-1 Internal and external co rrosion mechanisms in a subsea oil and gas production environment.

Corrosion
Mechanism
External Internal Chemical reaction Time
dependency
O2-corrosion X X 2Fe + H2O + 3/2O2 → 2FeO(OH)(s)
(“rust”)

Time
dependent
CO2-corrosion 1)
(sweet corrosion)
NA X Fe + H2O + CO2 → FeCO3(s) + H2

Time
dependent
Microbiologically
induced corrosion
(MIC)
X X Fe + “bacteria related oxidant” → Fe2+

Time
dependent/
abrupt
nature
Sulphide stress
cracking (SSC)
(corrosion due to
H2S)
NA2) X 2H+ → H⋅(ads)
H⋅(ads) → H2(ads) (inhibited by H2S)
H⋅(ads) → H⋅(abs)

Abrupt
nature
1) Not anticipated on corrosion resistant alloys
2) Under certain conditions high levels of H2S might occur in the seabed, however, such condition is not anticipated to occur on the Norwegian shelf.

2.1.2.2 External corrosion
External corrosion is f or m ost subm erged equipm ent controlled by the use of an external corrosion coating and a cathodic protection (CP) system . The de sign of the CP system is dependent on the design life of the equipm ent and the type and quality of the external coating system in question.

Some subsea components may not be provided with a CP-system (e.g. for chains) or CP will not be efficient due to shielding (e.g. water filled hollow profiles). For carbon steel components a corrosion allowance (CA) m ust then be added. The CA that must be added will depe nd on the availability of oxygen ( oxidant). For areas w ith limited access of oxyge n, such as within hollow profiles of structural steel, the corrosion rate will be low and a moderate CA is tolerable, whereas for instance chains that are freely exposed to seawater, a higher corrosion rate must be accounted for. The CA is norm ally determined as a part of the design and is based on the specified design life of the component.

Certain corrosion resistant alloys (CRA’s) and titanium are resistant to seawater corrosion under North Sea ambient seawater conditions and can be used without cathodic protection.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

2.1.2.3 Hydrogen Induced Stress Cracking
Cathodic protection m ay be detrim ental for so me m aterials due to Hydrogen Induced Stress Cracking (HISC). Hydrogen Induced Stress Crack ing (HISC) is caused by a combination of load/stress and hydrogen e mbrittlement (HE) caused by the ingre ss of atomic hydrogen into the metal matrix formed at the steel surface due to the cathodic protection (CP).

High strength steel (SMYS > 500 MPa) and some corrosion resistance materials (13Cr-steel and duplex stainless steels) are susceptible to HISC , see /1, 2, 3, 4/. Soluti on annealed austenitic stainless steels and nickel based alloys are generally considered immune to HISC.

DNV RP B401 Sec. 5.5 /1/ gives recommendations with regards to materials maximum hardness level and the specif ied m inimum yield s trength f or saf e com binations with CP. Bolts in martensitic steels heat treated to SMYS up to 72 0 MPa and maximum hardness level of 350HV (ASTM A182 grade B7 and ASTM A320 grade L7) have docum ented com patibility with C P (see also Norsok M-001 Sec. 5.6 /5/).

Factors influencing HIS C of duplex stainless steel have been recap itulated in a draft version of DNV RP F-112 /2/ with recomm endation for desi gn criteria based on best practice and on today’s knowledge (strain/stress criteria).

HISC is abrupt of nature and it is expected to occur durin g the first y ears of the installations design life if the conditions are ideal.

2.1.2.4 Internal corrosion
CO2-corrosion; CO2-corrosion or sweet corrosion is not anticipated for corrosion resistant materials (e.g. 13Cr, 316L, 22Cr, 25Cr, Alloy 625). Carbon steel, however, will be subjected to CO2-corrosion. The corrosion rate is depe ndent on the partia l pressure of CO 2, the tem perature, the flow regime and the water in-situ pH. The c orrosion takes the form of localised- (‘pitting’), uniform- and grooving- (e.g. longitudinal, tran sverse) attacks and is a tim e dependent degradation m echanism. CO 2-corrosion can be m itigated by the use of corrosion inhibito rs and/or by pH- stabilisation of the process fluid (primarily applicable for pipelines).

O2-corrosion; Internal corrosion due to the presence of O 2 is in principle not expected in oil and gas production system s since no oxygen shall be present in the process m edium. Ingress of oxygen may increase the corrosion in the system.

Water used for water injection can be either deaerated or aerated, which will have an impact on the corrosivity. Due to the rem oval of oxygen in deaerated water, th e corrosion rate of carbon steel will b e low, whereas in sys tems carrying aera ted wa ter a h igher corros ion r ate m ust be anticipated. Oxygen corrosion is a time dependent corrosion mechanism and takes principally the form of uniform corrosion, but localised attacks may also occur (‘pitting’).

Corrosion resistance alloys (CRA’s) and titanium can be used for seawater service but there are certain design lim itations regardin g the use of such m aterials (e .g. tem perature, presence of crevices, ch lorination etc.). Co rrosion of CRA takes th e fo rm of localised attack. Unfortunate combination of m aterial and operating environ ment will for m ost cases result in a corros ion failure during the initial phase of an installation’s life.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Environmental cracking due to H 2S; Corrosion due to the presence of H 2S is primarily rela ted to environmental cracking (i.e. sulphide stress cracking (SSC)). Both carbon steel and CRA’s are susceptible to SSC. The risk for SSC is de pendent on the partial pressure of the H 2S, the in-situ pH-value, total tensile stress, chloride ion con centration, presence of other oxidant etc. (for details reference is m ade to ISO-15156). Below a critical partial pres sure of H 2S no SSC is expected to occur. However, for partial pressures above this limit there is an increasing risk for SSC and the environm ental condition is term ed as sour. The resu lting failure m ode is crackin g and it is of abrupt nature. SSC is controlled by specification of the m aterial pro perties (e.g. hardness) and the m anufacturing process. For sus ceptible materials, environmental cracking is expected to occur during the initial phase of production and is not expected to have a tim e dependent development similar to ‘sweet’ corrosion.

Older petroleum installations m ay experience a souring of the wells (t he produced am ount of H2S increases) and the production environm ent turns from sweet to sour. The risk for environmental cracking should for such cases be subjected to evaluations with respect to the material properties and the new service condition.

Microbiologically Induced Corrosio n (MIC); The two best known bacteria of concern for the oil and gas industry are the sulphate reducing bacteria (SRB) and the acid producing bacteria (APB). They m ay live synergis tically in colon ies attached to the steel surface, where the SRB bacteria live beneath the APB colony. SRB bact eria live in oxygen-free environm ents and use sulphate ions in the water as a source of oxyge n. H2S is pr oduced as a waste product from the SRB, producing a corrosive environm ent locally in connection with the colony of bacteria. The risk for obtaining MIC will d epend on the availability of nutrients, temperature, water and flow condition. MIC takes the form of localised attack causing a pinhole leakage of a pipe. High corrosion ra te c an be a nticipated ( >1 mm /year) if the co nditions a re idea l. M IC has b een obtained in oil production system s as well as on steel exposed (e .g. anchor chains) to seabed sediments. The loca tion of MIC is dif ficult to predict. F or pipe line system s, tre atments with biocide m ay be effective as a preventive m easure. A common source for bacteria in a closed system is seawater. Use of untreated seawater for hydro testing should therefore be avoided.

Galvanic corrosion; Galvanic corrosion may occur when there is an electrical coupling betwee n dissimilar metals. The least noble m aterial (an ode) will be sacrificed on behalf of the noblest material (cathode). The extent of accelerated corrosion resulting fro m galvanic coupling is affected by the electrochem ical potential difference between m etallic couple, the n ature of the environment (corrosivity) and the area ratio of anodic- and cathodic areas (sm all anode to cathode area ration is unfavourable).

Galvanic corrosion is a tim e dependent form of corrosion and resu lt in a uniform corros ion attack. The possibility for obtaining galvanic corrosion should be evaluated during the design phase. For cases where a galvanic couple is in evitable, a distance spools of a non-conducting material can be installed or in stallation of a galvanic spool w ith sufficient wall thickness where the material is intended to corrode (i.e. sacrificial spool).

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

2.1.3 Fatigue
Fatigue failures occur in parts which are subjected to alternating, or fluctuating (other used terms are dyn amic or cy clic), stres ses. Fatigue crack ing is u sually initia ted at str ess raise rs such as sharp geometric transitions, welds, notches or inte rnal material flaws such as slags, cracks (e.g. quench cracks or weld lack of fusion defects ). A m inute crack st arts at a l ocalized spot and gradually spreads over the cross section until th e component/member fails due to overloading of the remaining cross section area.

Fatigue results in an almost brittle-appearing fracture, with no gross d eformation at the f racture.

Crack propagation m ay be divided into stages , the m ost im portant being “initiation” and “propagation”. Depending on the mate rial, the stress level and eventual environmental impact - initiation o r propaga tion m ay be the stag e co nstituting th e m ain part of the lif etime of the component/member.

Fatigue failure is c aused by a critic al localized tensile stress which is v ery difficult to evalua te and therefore design for fatigue failure is ba sed prim arily on em pirical relationships using nominal stresses. A fatigue failure can usually be recognized from the appearance of the fracture surface, which shows a sm ooth region, due to the fatigue crack p ropagation through the sectio n (being more or less inse nsitive to the microstructure and he nce orientated perpendicular to th e principle direction of the appl ied stress), and a rough region, wh ere the member has failed when the remaining cross section was no longer able to carry the load. Frequen tly the progress of the fracture is indicated by a pattern of parallel lines, or “beach marks”, progressing inward from the point of initiation of the failure.

Three basic factors are necessary to cause fatigue failure. These are (1) a maximum tensile stress of sufficiently high value, (2) a large enough variation or fluctuation in the applied stress, and (3) a sufficiently large number of cycles of the applie d stress. In addition, there are a host of other variables, such as stress concen tration (e.g. geometric transitions), corrosion (localised corrosion e.g. preferential weld corrosion), temperature (stresses induced by linear therm al expansion or differences in thermal expansion coefficients between different materials), overload (over-torque of bolts), metallurgical structure (e.g. direction of texture vs. direction of applied stress), residual stresses (welding or form ing residual stresses), and combined stre sses, which tend to alter the conditions for fatigue.

Recommendations for design when considering fa tigue resistan ce of offshore structures are given in D NV-RP-C203 /6/. Th is recomm ended practice cont ains am ong other things a collection of fatigue resistance (S -N) curves developed for different configurations (e.g. types of welded joints), surface finish and env ironmental conditions. For some of the curves assumptions regarding defect sizes, and hence corresponding NDT requirements are given.

Crack propagation may be both adve rsely or favourably affected by variable am plitude loading.

Intermittent high s tresses m ay crea te large pla stic zones with “r esidual” compressive s tresses ahead of the crack toe and hence retard crack growth.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

2.1.4 Corrosion fatigue
On a general level fatigue is affected by environm ental conditions and in particular by corrosion.

HISC and sour service conditions, as desc ribed in Section 2.1.2.2 and 2.1.2.3, respectively, m ay facilitate fatigue crack initiation. M etal loss by corrosion will gen erally enhance crack growth, but under reversed loading condi tions (tension – co mpression) corrosion products m ay reduc e the “impact” of the total stress range due to “crack closure”.

Corrosion f atigue occu rs in m etals as a resu lt of the combined action of cyclic s tress and a corrosive environm ent. For a given m aterial, th e f atigue streng th (o r f atigue lif e at a given maximum stress value) generally decreases in the presence of an aggressive environment.

When corrosion and fatigue occur sim ultaneously, th e chem ical attack accelerates the rate at which fatigue cracks propagate. Materials which show a definite fatigue limit when tested in a ir at room te mperature show no indication of a fa tigue limit when the test is ca rried out in a corrosive environment.

Corrosion fatigue crack growth m ight be influen ced by m any variables, such as those listed in Section 2.1.3, but also by environm ental variables (gaseous or liquid environm ent, partial pressure of damaging species in gaseous environments, temperature, pH).

A number of methods are available for minimizing corrosion fatigue damage:
- The choice of material for this type of service shoul d be based on its corrosion resistant properties rather than the conve ntional fatigue proper ties (ex. stainless steel over heat-treated steel)
- Protection of the m etal from contact with the corrosive environm ent by m etallic or non-m etallic coatings (provided that the coating does not becom e ruptured from the cyclic strain)
- Addition of a corrosion inhibitor in closed systems to reduce the corrosive attack - Elimination of stress concentrators by careful design

2.2 References

/1/ DNV-RP-B401 (2005), Cathodic protection design
/2/ Draft-DNV RP-F112, Design of du plex stainl ess steel subsea equipment exposed to cathodic protection
/3/ Thierry Cassagne and Freddy Busschaert, ”A review on Hydrogen Embrittlement of duplex stainless steels”, NACE Paper 05098 cor
/4/ E. B. Heier and R. B. Johansen “North Sea Failures of 13Cr flowlines: Consequences for Future Application”, Proc. of SEM X, Costa Mesa, California, June 2004. /5/ Norsok M-001 rev. 4, Materials selection
/6/ DNV-RP-C203 (2005) Fatigue design of offshore steel structures

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
3 LOAD BEARING STRUCTURES

3.1 Concrete

3.1.1 Introduction
When the first concept of fixed concrete stru ctures for offshore oil and gas exploration and production in the North Sea was introduced in th e late 1960´s, the offshore technical community showed m uch scepticis m. At the sam e ti me, however, the results of a com prehensive field investigation of more than 200 existing conventional concrete sea st ructures such as bridges and harbour structures alo ng the Norwegian co astline were publish ed /1/. The overall goo d performance of these structures reco rded even af ter a serv ice life of 50-60 years contributed to convincing the technical community that also conc rete could be a reliable construction m aterial for oil and gas installations in the North Sea. However, the appearance of corrosion on embedded reinforcement steel that typically too k place on the conventional concrete sea structu res already after a service period of 5-10 years was not acceptable. Therefore, in order to gain acceptance for the f irst of fshore conc rete pla tform in the No rth Sea, both increased concrete qualities and concrete covers beyond the requi rements of current concrete codes were required. Secondly, much stricter prog rams for QA/QC com pared to the exis ting design an d construction practice had to be introduced.

Already during the con struction of Ekofisk Ta nk, the first edition of “Recommendations for design and construction of concrete sea st ructures” w as published by the international organization for prestressed concrete /2/. Ther eafter, both Norwegian Pe troleum Directorate in their Regulations /3/ and Det No rske Veritas /4/ in their Rule s had adopted new and stricter durability requirements for fixed offshore concrete structures.

After the first breakthrough for us e of concrete in developing the Ekofisk oil field, a rapid development took place. Duri ng the period from 1973 to 1995 alt ogether 28 m ajor concrete platforms containing more than 2.5 m illion cubic meters of concrete were installed in the North Sea. Several of these in stallations are now succe ssively approaching the in tended service life of 25-30 years.

Considering the harsh and hostile m arine environment in the North Sea, the question has been raised on ho w these con crete structures have performed so far. As th ere is need for a continu ed service of the installations beyond the service life originally designed, the question also has been raised for how long these fixed concrete structures can be safely operated.

3.1.2 Main degradation mechanisms
Although deteriorating mechanisms such as chemical seawater attack, freezing and thawing, and expansive alkali reactions all present som e pot ential durability problem s for offshore concrete structures, it is relativ ely easy to avo id such problems by taking the neces sary precautions at an early stage of planning, designi ng and construction. For oil cont ainment vessels, aggressive bacteriological environment may also represent a potential problem of concrete degradation. For a dense, high-quality concrete, however, such a degradation should not represent any durability problem.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
The concrete platform s located in the Norwegia n Sector of the North Sea were designed and constructed in acco rdance with th e requirements specified in /2, 3, 4/, and reflected the current “state-of-the-art” which in the main are still relevant and acceptable.

For concrete structures in the m arine environment, extensive experience demonstrates, however, that it is not the disintegration of the concrete itself but rather the electro chemical corrosion of embedded s teel reinforcem ent and pre-stressed tendons which poses the m ost critical and greatest threat to the dur ability and long-term performance of th e structures /5/. As long as it is possible to prevent or retard th e chlorides penetration into concrete, all em bedded steel is very efficiently protected from corro sion by electrochem ical passivatio n of the highly alkaline concrete.

The high alkalinity of concrete may be neutralized by a reaction between the atmospheric carbon dioxide and the calcium hydroxide solution in the concrete. Such a carbonation process will also impair the passivity of embedded steel causing steel corrosion. For a dense, high-quality concrete in a moist environment, however, carbonation is limited to a very thin surface layer and does not represent real problem to marine concrete structures.

As soon as the chlorides from seawater have reached embedded reinforcement and prestressed steel and th e passivity of the ste el is broken, a complex system of galvanic c ells will dev elop causing electrochemical corrosion of the steel /6/. The rate of embedded steel corrosion will then primarily be controlled by the availability of dissolved oxygen in the cathodic areas and the electrical resistivity of the concrete. The area ratio of th e depassivated part s (anodic areas) and the pass ive parts (cath odic areas) in the galvanic c ells is also an im portant factor for the corrosion rates. The electrical resistivity of concrete in moist marine environment is normally so low that it does not become the governing factor for the electrochemical corrosion process /7/.

Both in the tidal and in the spla sh zones of concrete sea structures, oxygen is available in plenty, so that a hig h corrosion rate can tak e place in th e zones /8/. Only for the constantly subm erged parts of hig h quality co ncrete structures, the av ailability of oxygen is g enerally so low that an electrochemical corrosion of embedded steel does not represent any practical problem /9/.

In all concrete structures, a certain am ount of c racks in the concrete cover m ay freely expose parts of the em bedded steel. In a moist m arine environm ent, however, crack widths of up to

0.5 mm would norm ally not represent any corrosi on problem s /10/. For static cracks in submerged parts of the structure, even wider cr acks may be tolerated due to a filling up of the cracks by various chem ical reaction products (Cal careous depositions consisting m ainly of magnesium hydroxide, calcium carbonate and corrosi on products which block up the cracks and reduce rate of steel corrosion. The process is kn own in the literature as “self-healing of cracks”).

For dynamic cracks, however, wider (in excess of 0.5 – 0.6 mm) cracks may represent a potential corrosion problem /11/.

In the subm erged part of offshore concrete st ructures, a variety of freely exposed m etallic components, such as pipes, penetration sleeves, clamps, brackets, supports and other fixtures are in m etallic connection with th e embedded reinforcem ent steel and m ay represent a special corrosion problem /12/. In such a case, the fr eely exposed m etallic components represent sm all anodic areas in m etallic connection with huge cathodic areas of the embedded reinforcem ent steel acting as catchm ents areas for oxygen. In orde r to control this galvanic corrosion problem , an effective cathodic protection sys tem for all fr eely exposed metallic components is ess ential.

Normally, such a cathodic protection has been based on sacrificial anodes.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

3.1.3 History
Extensive surveying programs both above and below water are regularly being carried out for all offshore concrete structures in the North S ea, however no recent inform ation on the curren t status of these structures is available for gene ral public. In a State of the Art report from 1994 /13/, very little corrosion problem s on e mbedded steel was reported ; the m ost serious problems were related to accidental loads from ships and falling items. More detailed inspections of eleven of the oldest concrete structures instal led during 1973–78 and reported in 1982, also showed their generally good co nditions /14/. Inspectio n of the concrete platform s on the Statfjord and Gullfaks Oil Fields in 1992 also reported a generally good condition of the concrete /15/.

After 20 years of service, regular inspections of the relatively wide concrete cracks found at the foundation of Frigg CDP-1 Platform had revealed no serious corrosion in the cracked concrete /16/. After 18 years of service, corrosion m onitoring based on embedded steel tubes in the F rigg TCP-2 Platform did neither report any steel corrosion /16/. Also laboratory-based investigations have shown that the risk of steel corrosion in cracked sub merged concrete appears to be less severe than what originally expected /10, 11/.

Only f or a f ew of the concre te pla tforms in the North Sea, m ore system atic investigations of chloride penetration have been carried out. All of these investigations clearly demonstrate that a certain rate of chloride penetra tion does take p lace, but only at a slower rate com pared to that typically observed on conventi onal concrete structures /17/. Figure 3-1-1 and 3-1-2 show the chloride penetration into the concr ete of Statfjord A Platform and Ekofisk Tank after 8 and 17 years of exposure, respectively. Figure 3-1-3 and 3-1-4 show the chloride penetration into the concrete of Brent B and Brent C Platform s, re spectively, after 18-20 years of exposure. For Oseberg A Platform , where th e co ncrete cover wa s partly less than w hat had been specified, serious corrosion problem s occurre d already at an early stage of operation. For this structure, repairs in the form of cathodic protection were carried out in 1998 /21/.

Figure 3-1-1 Chloride penetration into con crete of Sta tfjord A Platform (19 77) after 8 years of exposure /18/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 3-1-2 Chloride penetration into con crete of Ekofisk Tank (1973) after 17 years of exposure /19/.

0 0,05
0,1
0,15
0,2
0,25
0,3
0,35
0 102 03 0 405 0 607 0
Distance from the exposed surface (mm)
Chloride concentration
(% Cl as weight of concrete)
elevation: + 14.4 m
elevation: + 7.83 m
elevation: -11.5 m

Figure 3-1-3 Chloride penetratio n into concrete of Brent B Platform (1975) after 20 years of exposure /20/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 3-1-4 Chloride penetratio n into concrete of Brent C Platform (1976) a fter 18 years of exposure /17/.

For Statfjord A Platform the design specification re quired that the shafts of the structure should be protected in the splash zone by an epoxy coating. From Figure 3-1-1 it can be seen that such a coating had very efficiently prevented the chlorides from penetrating the concrete, and even after 15 years, this protection still appeared to be very effective /22/. However, for Heidrun Platform, Figure 3-1-5 shows that a much poorer surface coating partly applied to the legs of th is platform was not so efficient in keeping the chlorides out, even after an exposure period of only two years.

Figure 3-1 -5 Effect o f concrete co ating on ch loride penetration into concrete o f He idrun Platform (1995) after 2 years of exposure /17/.

For all concrete structu res in the North Sea, a minimum concrete cover of 75 mm for the splash zone was used. Although very strict p rograms for QA/QC also were im plemented during concrete construction, it is a typi cal feature of all concrete struct ures that both concrete covers and concrete qualities always show a high scatter and v ariability /23 /. In spite of the lim ited Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
information available on chloride penetration into the concre te structures in the North Sea, the general conditions appear to be very good.

However, for several of the structures, a certa in am ount of steel corrosion has already been observed and m inor repairs carried out. The above results indicate th at the chlorides m ay penetrate the specified concrete cov er in the splash zone within a service period of 25-30 years.

For the Brent C Platform (1975), where enough data was available to carry out a probability- based durability analysis, Figure 3.1.6 shows that a risk level of 10 % for steel corrosion is rapidly exceeded after a service period of approximately 20 years.

0 10
20 30
40 50
60 70
80 90
100
02 0 40 60 80 1
Time (years)
Risk for depassivation (%
00 )

Figure 3-1-6 Development of risk for steel corrosion in the Brent B Platform (1975) /20/.

In addition to the corrosion of e mbedded steel, the potential corrosion problem for all the f reely exposed metallic components attached to the concrete structures has already been pointed out.

For majority of the platforms, a relatively high anode consumption at an early stage of operation has been observed, but after some tim e, the ra te of anode consumption has typically been reduced to a m ore stable value. Current experience w ith rates of anode c onsumption appears to vary f rom one location to another within the sa me structure and als o from one structu re to another /24/.

Since different guidelines and recommended practice for cathodic protection of the structures have been used for the design of cathodic protection system , different design values for current drainage to the em bedded steel were applied /24/. It is im portant, therefore, that the inspection schedule and procedure are relevant for a particular stru cture in order to ensure a close and systematic monitoring of the anode consumption rates.

3.1.4 State of the Art on R&D
For all th e concrete structures in the North S ea, the durability specifications were based on prescriptive requirem ents on th e composition of concrete m ixes and on execution of concrete work. In order to obtain a m ore controlled durability and service life of ne w concrete structures

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
in marine environment, a rapid d evelopment of probability -based durability design has rec ently taken place /25/. Durability des ign based on thes e new principles has now becom e the basis f or new recommendations and guidelin es for increasing durability of Norwegian concrete harbou r structures /26, 27/. These new design procedures include the following elements:
- Probability-based durability analysis
- Evaluation of alternative strategies and protective measures - Documentation of obtained construction quality and durability - Preparation of a service manual for regular condition assessment and monitoring of chloride penetration with protective measures for control of this penetration.

For the exis ting conc rete struc tures in the Nor th Sea, if regular observations on the rates of chloride penetration into the structures are made, the same probability-based procedures can also be applied f or a m ore reliable extr apolation of the furthe r chloride penetration and risk of steel corrosion.

3.1.5 Recommendations
As several of the concrete structures in the North Sea are no w approaching the inten ded service life of 25-30 years, current inform ation indica tes that an increasing amount of c orrosion on embedded s teel m ay be expected in the years to com e. For increased service periods beyond what was originally specified, this may represent a future challenge to the operators. In order to better meet this challenge, a closer following up of the rates of chloride penetration in the splash zone of the structures is recomm ended. This following up should include regular m easurements of chloride penetration in give n critical lo cations of the shafts . Based on such m easurements, numerical procedures for a probabilistic extrapol ation of the further chloride penetration and a risk of e mbedded steel corrosion are available. More data on the chloride penetration into concrete is provided the more accurate and reliable such an extrapolation will be /24/.

In order to avoid unnece ssary galvanic corrosion problem s on the f reely expo sed m etallic components attached to the concrete structures , a proper monitoring of the sacrificial anode systems for these components is of vital importance.

3.1.6 References

1. Gjørv, O.E., ”Durability of Reinforced C oncrete W harves in Norwegian Harbours”, Ingeniørforlaget AS, Oslo, 1968, 208 p.

2. Féderation Internationale de la Précontrain te (FIP), “Recommendations for the Design and Construction of Concrete Sea Structures”, London, 1973.

3. The Norwegian Petroleum Directo rate, “Regul ations for the Stru ctural Design of Fixed Offshore Structures”, Stavanger, 1976.

4. Det Norske Veritas, “Rules for the Design, Construction and Inspection of Fixed Offshore Structures”, Oslo, 1976.

5. Gjørv, O.E., ”Perform ance and Serviceability of Concrete Structures in the Marine Environment”, Proceedings, Odd E. Gjørv Sym posium on Concrete for Marine S tructures, ed. by P.K. Mehta, CANMET/ACI, Ottawa, Canada, 1996, pp. 259-279.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

6. Gjørv, O.E., “Mechanisms of Steel Corrosion in Concrete Structures”, Proceed ings, Vol. 2, EVALMAT 89, International Co nference on Evaluation of Ma terials Perform ance in Severe Environments, Kobe, Japan, 1989, pp. 565-578.

7. Gjørv, O.E., Vennesland, Ø. and El-Busaidy, A.H. S., “Electrical Resistivity of Concrete in the Oceans”, Offshore Technology Conference, Houston, Texas, Paper No. 2803, 1977, pp. 581-588.

8. Gjørv, O.E., Vennesland, Ø. and El-Busaidy, A.H.S., ” Diffusion of Dissolved Oxygen through Concrete”, Materials Performance, Vol. 24, 1986, pp. 39-34.

9. Gjørv, O.E. and Kashino, N., “Durability of a 60 Year Old Reinforced Concrete Pier in Osl o Harbour”, Materials Performance, Vol. 25, 1986, No. 2, pp. 18-26.

10. Vennesland, Ø. and Gjørv, O.E., 1981, “Effe ct of Cracks in Subm erged Concrete Sea Structures on Steel Corrosion”, Materials Performance, Vol. 20, 1981, pp. 49-51.

11. Espelid, B. and Nilsen, N., “A Field Study of the Corrosion Behavior on Dynam ically Loaded Marine Concrete Structures”, ACI SP-109, ed. by V.M. Malhotra, 1988, pp. 85- 104.

12. Gjørv, O.E., “Steel Corrosion in Offshore Concrete Platforms”, Proceedings of the Industrial Seminar “T he Ocean – Our Future”, The Norwegian University of T echnology, NTH, Trondheim, 1977, Tapir, Trondheim, 1977, pp. 390-399. (In Norwegian).

13. FIP Concrete Sea Structures Comm ission, “Durability of Concrete Stru ctures in the North Sea”, 1994, State-of-the-Art-Report, London, 1994.

14. Fjeld, S. and Røland, B., “Experience from In -Service Inspection and Monitoring of 11 North Sea Structures“, Offshore Technology Conference, Paper No. 4358, Houston, USA, 1982, 9 p.

15. Hølaas, H., 1992, “Condition of the Concrete St ructures at the Statfjo rd and Gullfaks Oil Fields”, Report OD 92/87, The Norwegian Pe troleum Directorate, Stavanger, (In Norwegian).

16. Elf Petroleum Norge A.S, (Private communication in 1995).

17. Gjørv, O.E., “Durability and Service Life of Co ncrete Structures”, Proceedings, The First fib Congress 2002, Session 8, V. 6, Japan Prestresse d Concrete Engineering Association, Tokyo, 2002, pp. 1-16

18. Sandvik, M. and Wick, S.O., “Chloride Penetration into Concrete Platforms in the North Sea “, Proceedings, Workshop on Chloride Penetrat ion into Concrete Structures, Division of Building Materials, Chalm ers University of Technology, Gothenburg, ed. by L.-O.

Nilsson, Gothenburg, 1993, 7 p.

19. Sandvik, M., Haug, A.K. and Erlien, O, “Chlor ide Permeability of High-Streng th Concrete Platforms in the North Sea “, ACI SP-145, ed. V.M. Malhotra, Detroit, 1994, pp. 121-130.

20. Department of Structural Engineering, Norw egian University of Science and Technology, NTNU, Trondheim. (Unpublished results)

21. Østmoen, T., “Field Tests with Cathodic Protection on the Oseberg A Platform “, Ingeniørnytt, Vol. 34, No. 6, 1998, pp. 16-17. (In Norwegian)

22. Aarstein, R., Rindarøy, O.E., Liodden, O. an d Jenssen, B.W ., “Effect of Coatings on Chloride P enetration into Offshore C oncrete Structures“, Proceedings, Second International Conference on Concrete unde r Severe Con ditions – Environm ent and Loading, ed. by O.E. Gjørv, K. Sakai and N. Banthia, E & FN Spon, London, New York, ISBN 0 419 23850 6, 1998, pp. 921-929.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

23. Gjørv, O.E., “Durability of Concrete Struc tures and Perf ormance-Based Quality C ontrol”, Proceedings, Internation al Conferen ce on Perfo rmance of Constructio n Materials in The New Millen ium, ed. by A.S. El-Dieb, M.M.R. Taha and S.L. Lissel, Sham s University , Cairo, ISBN 977 237 191, 2003, 10 p.

24. Ferreira, M ., Årskog, V. and Gjørv, O.E., “P robability-Based Durability Analysis of Concrete H arbour Stru ctures”, Pro ceedings, Vol. 1, Fourth Internatio nal Conference on Concrete Under Severe Conditions – Environm ent and Loading, ed. by B.H. Oh, K. Sakai, O.E. Gjørv and N. Banthia, Seoul, Korea Concrete Inst itute, ISBN 89 89499 02 X 93530, 2004, pp. 999-1006.

25. Gjørv, O.E., “Durability Design o f Concrete Structures”, Proceed ings, Vol. 1, Fourth International Conference on Concrete U nder Severe Conditions – E nvironment and Loading, ed. by B.H. Oh, K. Sakai, O.E. Gjør v and N. Banthia, Seoul, Korea Concrete Institute, ISBN 89 89499 02 X 93530, 2004, pp. 44-55.

26. “Recommended Specifications for Increased Durability of New Concrete Harbour Structures”, The Norwegian Association for H arbour Engineers, Oslo, 2004, 17 p. (In Norwegian)

27. “Recommended Specifications for Increased Durability of New Concrete Harbour Structures”, The Norwegian Association for H arbour Engineers, Oslo, 2004, 17 p. (In Norwegian)

28. “Practical Guidelines for Service Life Desi gn of New Concrete Harbo ur Structure s”, The Norwegian Association for Harbour Engineers, Oslo, 2004, 43 p. (In Norwegian)

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

3.2 Steel

3.2.1 Introduction
DNV has be en involved in class ification of offs hore units, mainly Column Stabilis ed Units an d Jack-ups since the 70’ties and 80’ties, and in the last two decades also ship-shaped units (FPSO’s and drilling vessels). T hrough this classificati on activity DNV has gained a wide experience related to degradation mechanisms for floating structures built in steel.

Normal design lif e for classed un its is 20 year s. The units built in the early 70’ties have now reached an age of approx. 30 years, which is si gnificantly m ore than they were originally designed for. Therefore, during the las t 5-10 years an increas e focus has been m ade to degradation m echanisms relevan t for these typ es of units, and how they can be dealt with in particular for units exceeding the original design life.

3.2.1.1 Main principles of classification
The effect of degradatio n mechanisms on the safety level of floating struct ures is closely linked to the principles and survey sc heme of classification. Classif ication is based on a renewal of the class certificate ev ery 5th year, see /1 /. This rene wal includes a detailed survey, which includes general visual inspection, close visual inspection in way of e xpected critical details, and non- destructive testing (NDT) accord ing to a pre -defined In-Service Ins pection P rogramme (IIP) prepared by the classification society (DNV). T he IIP is updated if experience for sim ilar units indicates problem areas not known and not covered by the IIP.

In addition to the major renewal survey every 5th year, an annual survey is also carried out each year, and intermediate survey in the middle of a 5 year period to maintain the validity of the class certificate.

All together the classification su rvey scheme has proven to be a good tool to control the m ost common de gradation mechanism s for floating units m ade of steel. The relevant degradation mechanisms are discussed in more detail below.

3.2.2 History
There have been relativ ely few m ajor accidents due to degradation of floating units. The m ost severe accident is Alexander Kielland (1980), which lost one colum n due to fatigue cracking.

The crack started from welded detail in way of a hydrophone (Figure 3-2-1 and Figure 3-2-2). At that tim e th e brac es we re norm ally f illed with water, whic h m ade it impossible to detec t an y leakage due to the fatigue crackin g, and the crack could grow to a critica l siz e, with the following rapture of the brace element.

Other minor incidents have occurred, but in genera l it is f air to say tha t the d egradation of the units have been monitored and controlled through the inspection programs, and necessary action such as m aintenance and repair has been carried out to m aintain the ov erall safety of the units avoiding major accidents.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 3-2-1 Alexander L. Kielland – loss of column (Source: T he Alexander Kielland Accident. NOU 1981:11) /2/.

Figure 3-2-2 Alexand er L. Kiella nd – positi on of cra ck (Source: The Alexander Kielland Accident. NOU 1981:11) /2/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

3.2.3 Main degradation mechanisms
There are two main degradation mechanisms relevant for the overall integrity of an offshore steel structure:
- Corrosion: uniform/pitting (see Section 2.1.2.3, O2-corrosion) - Fatigue (see Section 2.1.3)

3.2.3.1 Corrosion
In general corrosion is a visual degradation m echanism which can be monitored by scheduled inspection e.g. as specified by the classifica tion schem e. Experience from 35 years with classification of floating offshor e units is tha t corros ion is m ainly a m echanism which caus es local structural damage, but no major reduction in the overall safety level of the units as long as proper inspection schemes and maintenance is provided.

Already from the early 70’ties when the classifi cation activity started, DNV had relatively strict requirements to corrosion protection. DNV introdu ced specific requirement to the quality of the corrosion protection arrangem ent. Most common is coating and sacrificial anodes, and to some extent im pressed current. Most ship-shaped u nits are als o built with thickn ess a llowance to account for corrosion. Column-stabilised units and jack-ups are normally not built with thickness allowance, since coating is required in all corrosion critical areas.

3.2.3.2 Present situation – Ageing rigs
Due to the Owner’s m aintenance schemes and also the Rule requirem ents applied already from the 70’ties, most floating units following these Rules are in gene ral in good condition. The m ain problem areas have be en inte rnal ta nks used f or tr imming of the unit. Such tanks will have a frequency of filling/emptying, which gives go od conditio ns for corro sion (Figure 3-2-3 and Figure 3-2-4). Such areas are g iven special focus in the class inspections. In general corrosion is not found t o be a safe ty critical degradation mechanism provided a classification in-service inspection programme is followed and proper actions are taken based on the findings. At each renewal survey DNV checks that the rig has an acceptable condition for 5 new years in operation. The acceptance criteria are the same for new and old units and therefore the age of the unit is not important as long as necessary inspection and maintenance is carried out, see /1/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 3-2-3 General Corrosion inside ballast tank (DNV photo).

Figure 3-2-4 Heavy Corrosion inside aft peak ballast tank.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

3.2.3.3 Fatigue
Fatigue (Section 2.1.3) has been the m ost important and most focused degradation mechanism for floating offshore structures. Th ere are sever al challenges related to f atigue as a degrada tion mechanism and as a design parameter for ageing rigs.

Analysis Methodology
Floating units built 20-30 years ago were designed based on the methodology and design requirement valid at that tim e. Lim ited experience was available, and software and com puter capacity lim ited the possibility to carry out detailed analysis for critical areas. Today the computer capacities allow more detailed fatigue analysis to be carried out. Such analysis together with experience (crack history) from operation of these un its the last 20-30 years have revealed that the fatigue life was over-estimated for some parts of the units. This means that some details, normally lo cal ho t-spots in way of bracket toes etc., have fatigue lif e less th an the requ ired design life when re-calculated according the present methodology.

Load history
Floating mobile units are normally designed to operate world-wide based on the scatter diagram for the North Atlantic, assum ing an equal dist ribution of the wave di rection. T he approach assumes that the unit is moved around with dominating waves from all direction. In reality some units may stay on one location for a long period w ith one dominating wave direction. This m ay cause increased fatigue damage in some areas, while other areas are less loaded.

Fabrication
The fabrication quality of fatigue critical details is of vital importance f or the fatigue capacity.

Local workm anship and com pliance with desi gn drawings are both im portant. The design calculations are based on a define d quality of the workmanship and also based on structural details as given on design draw ings. Poor workm anship (e.g. substandard welding) or wrong details (e.g. details not b uilt in accordance with design drawings) may reduce the actual fatigue capacity significantly. The crack causing the Alexander L. Kielland accident starts from a detail in way of the hydrophone which was not shown on the design drawing.

Rule development
After the Alexander L. Kielland accident the rule requirements were changed. The main changes related to fatigue and consequences of fatigue were as follows:
- all braces (referred to as slender members) shall be watertight and redundant - water leakage detection to be installed in all braces to detect water leakage due to fatigue cracking in an early stage
- additional damage stability requirement
These new requirement focused on the rigs possibili ty to survive after an accident / damage, and also to detect a crack p ropagation as early as possible. Early de tection gives tim e to p lan an implement repairs or other compensating measures.

Present situation - Ageing rigs
DNV has introduced a Fatigue Utilisation Index (F UI) as a param eter to m easure the “used” fatigue life. The param eter takes into account the number of years in operation and where it has operated.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
life fatigueDesign
life fatigue Used=FUI
More detailed definition is given in DNV-OSS-101, Ch.3 Sec. 1, I100.

Many floating units have reached their do cumented fatigue life (F UI>1) and are still in operation. S ome years ago DNV and the PSA star ted to focus on this situation, and DNV introduced s ome additio nal Rule re quirement for units ex ceeding the docum ented f atigue lif e.

These requirements are describe d in the DNV-OSS-101, Ch.3 Sec. 1 I. The content of this new requirements are as follows:

When a floating offshore unit reach its docum ented fatigue life one of the following options can be selected:

1. Units with no fatigue cracks during the first 20-30 years of operation can continue to follow the existing inspection schedule. Risk-based inspection m ethods show that unit with no cracks maintain the safety level also after exceeding th e document fatigue life as long as n o cracks are detected, see Figure 3-2-5. Th e m ethodology is described in OTC 11950 (2000): “Fatigue Reliability of Old Semi-submersibles”, /3/.

Fatigue life = 20 years
1,0E-04
1,0E-03
1,0E-02
1,0E-01
1,0E+00
0 5 10 15 20 25 30
time (years)
Failure probability
No Inspection
MPI every 5 year

Figure 3-2-5 Failure probability.

2. For units with fatigue cracks during first 20-30 years of operation one of the following steps should be taken, as decided by the Owner of the unit:

Owner shall assess s tructural deta ils in f atigue critica l areas with the purpose of improving the fatigue properties of the st ructure. The im provement m ay include, grinding, replacem ent of steel, m odification of details, implem entation of risk-based

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
inspection m ethods considering th e detailed information available fo r each detail, or a combination of these actions. In spection prog rammes will be updated to reflect the outcome of these investigations

or
The NDT requirem ents sim ilar to the exten t fo r ren ewal s urvey a re c arried out o n the intermediate survey. Th is means that the interval between the main NDT inspec tions is

2.5 years instead of 5 years. This is in lin e with results found from Risk-base Inspection methods, assuming that most of these units have had some fatigue cracks within the first 20 years of operation.

Similar requirement have been made for column-stabilised units and jack-ups.

The following inspection methods are involved for classed units in operation:
- visual inspection - overall inspection
- close visual inspection – inspection of pre-defined details expected to be critical - Non-Destructive Testing (NDT)
o Magnetic Particle Inspection (MPI)
o Ultrasonic Inspection (UT)
o Eddy Current (EC)
Scope and extent of inspection and NDT is given in the In-Service Inspection Programs prepared for each unit classed by DNV.

3.3 References
/1/ Rules for Classification of Offsho re Drilling and Support Units, DNV-OSS-101, October 2003.
/2/ The Alexander Kielland Accident. NOU 1981:11.
/3/ Fatigue Reliability of O ld Semisubmersibles, OTC11950, G. Sigur dsson, I. Lotsberg, T. Myhre and K. Orbeck-Nilssen, Det Norske Veritas, 2000

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
4 SUBSEA PIPELINES

4.1 Introduction
As pipelines become older, the pipeline operators have several new challenge s to consider, such as
– Changes in integrity, e.g. tim e dependent degradation m echanisms such as corrosion and fatigue (Section 2), or random mechanical damages (e.g. third party damages).
– Changes in inf rastructure f rom the as built, e. g. increas ed f ishing activity or heavier trawler gear.
– Changes in operational conditions, either as a natural change in well-stream condition, tie-in to other pipeline system or increased production rates. – Required to operate beyond the design lifetime.
– Design no longer valid due to the above mentioned issues.

4.2 History
Generally, review and analysis of historically causes of pipeline failures worldwide /1, 2, 3/, indicate tha t corrosion, specifically internal corrosion, is the m ost wi dely reported cause of failure for offshore pipelines, follow ed by maritime activities (e.g. anchor- or trawling- dam age and vessel c ollisions, so called third party dam age (TPD)) a nd then natural forces (e.g. storms and mudslides).

In the PARLOC 2001-report /1/, which includes a total number of 542 reported pipeline/riser incidents in the North Sea (at th e end of 2000), it is em phasised that there is a general opinion that the incident frequencies are highest in the early years of a pipeline’s life and towards the end of its life. The former has been attributed to hig her vessel activity during the first years of field development and/or early appearance of flaws re lated to design, m aterial, corrosion inhibition system etc. The latter is more related to changes in infrastructure from the as built, e.g. increased fishing activity or heavier trawler gear and corrosion of the system over time.

In the Nor th Sea, the o ldest pipeline is th e 36” Ekofisk-Teeside oil expo rt pipeline followed by the 36” Ekofisk-Em den gas export pipeline which cam e on stream in October 1975 and September 1977, respectively. Both pipelines are made of norm alised CM n-steel API 5L X60 (similar to SMYS 415). One of th e challenges related to the fu ture operation of the Em den pipeline has been that the pipeline was designed and installed prior to the first issue of the NACE MR-0175. Since the H2S content is forecasted to increase in the future, and with that the risk for sulphide stress cracking (Section 2. 1.2.3) if water is present, it ha s been of grea t importance to establish w hether the p ipeline m aterial is su itable f or a gr adual “transition” to “s our serv ice” condition or not.

Further, both pipelines are coat ed with an asphalt enam el coa ting but without reinforcem ent, which was not comm on before in the early 80’ ies. These old type of coating has shown tendencies to spall with age. The same problem is not reported for pipelines covered with asphalt

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
enamel including reinforcem ent or for polypr opylene coating (PP). The PARLOC 2001 report does not contain any information about coating types and type of coating damages.

The m ost comm only used m aterials for pipeline in the North Sea have been the carbon steel material grades X46, X52, X60 /3/. Later on more high strength steel as X65 and X70, and corrosion resistant steel as duplex/super duplex- and 13%Cr stainless steels, have been utilised as linepipe m aterial. I t has also during the last f ew years bee n installed quite a f ew CMn steel pipelines internally lined ( mechanical bonding) or clad (m etallurgical bonding) with CRA (e.g.

AISI 316L, Incoloy 825 and Inconel 625).

4.3 Main degradation mechanisms
Threats to the pipelin e system shall be syste matically id entified, ass essed and docum ented throughout the operational lifetime.

This shall be done for each section along the pipeline. Examples of typical threats are:

• corrosion (internal/external)
• third party damage (TPD)
• erosion
• development of free spans causing fatigue
• buckling

As a result of the natural aging of a pipeline, corrosion and third party damages are considered to constitute the most relevant threats to the system.

Internal Co rrosion i s t he m ost w idely r eported c ause of failure for subsea pipelines (see Section 4.2). The internal corrosion include s a large variety of corrosion degradation mechanisms depending on the process m edium, the material, the process condition etc. Internal corrosion in oil and gas pipeline s is princ ipally associated with CMn-s teel and th e f ollowing corrosion mechanisms (Section 2.1.2.3) are of main concern; - H2S-cracking (SSC)
- CO2-corrosion
- Microbiologically Influence Corrosion (MIC)
Liquid water is prerequisite for any electrochemical reaction causing corrosion to occur. Internal corrosion is controlled by m aterial s election (including clad pipe), applying a corrosion allowance (CA) on the inner surface of the C Mn-steel p ipe or by ch emical treatm ent of the process fluid (e.g. corrosion inhibitor, pH-stabilisation).

External corrosion is controlled by the use of a n external corrosion coating in combination with a cathodic protection (C P) system in case of coating damages (Section 2.1.2.2). The control of external corrosion will theref ore depend on the type and qua lity of the e xternal coating and the design of the CP-system.
• External co ating; Corrosion protection often consists of a tight protective layer around the pipeline exterior. T he extern al protective co ating is often asphalt enam el or fusion

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
bonded epoxy (FBE) covered with other types of plastics, as polyethylene or polypropylene, for m echanical protection or as heat insulation. Asphalt can only be utilised together with concrete coating (e.g. weight coating). • Concrete weight coating; Concrete is applied to the coated pipeline to provide the required compaction and density. T he thickness of the concrete ensu res both mechanical protection and density for negative buoyancy. Concrete weight coating system s provide the following advantages; sub-sea stability, pr evention of damage by e.g. ship' s anchors, no trenching and less steel is required.

With the exception of the coa ting system s used on the oldest p ipelines in th e North Se a (Section 4.2), no coating dam ages related to “ natural ag ing” of the coating itself have been reported /4/. The f ield joint coa ting (FJC), whic h are sp ecifically high risk areas fo r increased bacterial activity and a subsequent increase in the external corrosion, has been a topic for debate.

To DNV’s knowledge, no incidents have been re ported where significant wall thickness loss has been detected.

The quality and properties of m odern co atings and the q uality con trol asso ciated with the manufacturing of coating have been cons iderably im proved ove r the last yea rs.

Recommendations for the process of applying speci fic types of FJC/coating field repair (CFR) and 'infill' systems, and for the process of applying external coating systems for corrosion control of submarine pipelines at the coating plant are, as an exam ple, given in DNV-RP-F102 /5/ and DNV-RP-F106 /6/, respectively. As a consequence of the quality im provement, a considerably less conse rvative CP-de sign is ne cessary com pared with recomm endations in older CP-design codes (i.e. with respect to co ating breakdown factors and hence the net galvanic anode m ass to be installed). This is reflected in the ne wer CP-design codes, DNV RP-F103 (2002) and ISO 15589-2 (2004) /7, 8/.

Most experienced coating damages are related to external impact (TPD). The risk for third party damage (i.e. mechanical damage of the pipeline) is an issue during the entire life of the pipeline, however, as mentioned in Section 4.2, the incident frequencies are highest in the early years of a pipeline’s life and towards the end of is life.

All subsea system s (e.g. structures, pipelines, plat forms) shall principally be provided with its own CP-system . Intera ction in ter ms of curre nt drain between the pipeline CP system and adjacent subsea installations elect rically connected to e.g. p latforms or crossing pipelines m ay cause excessive anode consum ption of one of the structures. As the utm ost consequence a reduced design life of the CP-system and thereby an under-protection of the pipeline system may occur.

4.4 Failure modes
The main failure modes for pipelines are normally considered to be; • Leakage
• Burst
• Local Buckling / Collapse

Leakage in pipelines is often a ssociated with the pr esence of local corrosi on attacks (e.g. local CO2-corrosion, pitting), but might also be a result of small cracks. Burst is m ore associated with Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
a uniform wall thickness reduction or m ore extensive crack propagation, decreasing the pressure capacity of the pipeline. Local buc kling is a failure m ode confined to a short length of the pipe causing gross changes of the cross section; collapse. For subsea pipelines, this is often related to external overpressure in com bination with a wall thickness re duction (e.g. as a result of corrosion).

4.5 Ensuring integrity of subsea pipelines
For a given design, co rrosion monitoring, corrosion mitigating measures and inspection of the system are fundam ental activities to control the integrity of a pi peline system during its design life. As illustrated in Figure 4-1, several activities are necessary to be able to control the integrity of the pipeline system.

To be able to perf orm an integrity assessment of a pipeline system, the data and results from the activities illustra ted in the f igure ha s to be m ade availab le. One of th e challenges with older pipeline system is that historic al data and also often original design, fabrication and installation data and rep orts are lack ing. This complicates the possibility of perfor ming a reliab le integrity assessment.

Figure 4-1 Activities necessary to control the integrity of the pipeline system.

Condition
Monitoring
Process-, Product-
Control
Corrosion
Monitor.

External
Inspection
Internal
Inspection
Process
Data
Product
Analyses
Chemical
Injection

Pipeline Segment
Integrity Assessment
Corrosion
Assessment
Mechanical
Assessment

Pipeline Condition
Repair /
Modification

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
A short des cription of the m ain activitie s tha t constitute the pipe line integr ity m anagement system is given below. More detailed requireme nts to m anaging system integrity are given in industry standards as API 1160, ASMEB31.8S and DNV-OS-F101 /9, 10, 11/.

4.5.1 Process- and product control
Process- and product-control includes the following;
• Process control (pressure, temperature, flow rate etc) • Product sampling (CO2, H2S, O2, water cut/dew point, sand etc) • Chemical injection for corrosion prevention (corrosion inhibitors, pH-stabilisation etc.)
The process - and product contro l s hall ensu re th at the con dition in th e pipelin e is within the operational window as defined in design. If e.g. a souring of the well stream occurs, increasing the H2S-content and with that exceeding the m aximum specified lim it as given in design, this information should be handed over to the respon sible for pipeline integrity. For gas lines, the dew point should be m onitored and for liquid lines , the water cut should be known. Further, for chemical injection, the availability (or efficiency) given in design should be known together with the precau tions taken to ensure th e chosen av ailability. As an example, a 95% inhibitor availability requires, according to NORSOK M-001, that a qualified inhibitor is injected from day one and that a corrosion m anagement system is in plac e to a ctively monitor corrosion and inhibitor injection. Any redundancy in the system should be elucidated ensuring e.g. continuous injection of corrosion inhibitors even though one pum p fails and thereby m aintaining the required availability.

4.5.2 Corrosion Monitoring
The rate of corrosion dictates how long any pr ocess equipment can be safely operated. W hen applying corrosion m onitoring techniques it is vita l that the equipment is installed in locations where corrosion m ight occur (e.g. lo west points where liquid m ight accum ulate in confined areas). Othe rwise, the d ata re ceived f rom su ch m easurements will ha ve no relev ance when assessing the integrity of the system.

Typical corrosion monitoring techniques are;
• Corrosion coupons / ER-probes / LPR probes
• Sampling
• Field Signature Method - corrosion monitoring
• DDL - Deposition Data Logger
• UT of fixed spots
• Sand/Erosion

Off the techniques listed above, corrosion coupo ns, ER- and LPR-probes together with sam pling form the core of industrial m onitoring system s. The other techniques are norm ally found in specialised applications.

However, it should be em phasised that corro sion coupons and probes are not suitable for documenting the prospective corrosion in a pipe line but to m onitor any changes in the fluid corrosivity.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

4.5.3 External Inspections
Typical external inspection methods are;
• Visual Inspection (GVI/CVI) performed by divers
• Inspection perform ed by using Rem ote Op erated Vehicles (ROV); Video, Sonar, Transponders, Profilers, etc.
• External Ultrasonic Testing and Thickness Measurement for verification of metal loss or cracks
During such inspections the following is typically inspected:
• The CP system - looking for excessive consumption of the anode mass. • Indication of inadequate coverage of buried or rock dumped pipeline sections • Visual inspection of anode consumption
• Recording of anode potential and steel potential if practically • Field gradients at anodes
• Coating or concrete damages
• General dam age to pi pelines from i mpact (dropped object, equipm ent handling, anchor impact or dragging, fishing, etc.)
• Flanges and hubs – looking for leaks
• Looking for upheaval buckling or snaking

For buried pipelines, as-laid surveys along the entire length of the pipeline have to be performed prior to ba ckfilling (buried). Significant damages to the co ating and sa crificial anodes shall be recorded and the consequences for long-term performance consid ered. When the pipelines are buried, no further inspections of the coating or anodes are possible.

4.5.4 Internal Inspection
In-line inspections (ILI) are no rmally performed to v erify the in ternal surface condition of the pipeline system but are also, de pending on the chosen tool, capable of verifying the external condition with respect to corrosion and cracking.

Several types of tools for intern al inspection, cleaning and batchi ng are available on the m arket.

An overview of types of tools associated with in-line inspection (ILI) and cleaning/batching of pipeline-systems is given in Figure 4-1.

Additionally, ILI m ight be perform ed to m onitor the efficiency of the co rrosion protection (e.g. external coating) and th e prevention systems (i.e. corrosion inhibitor, de w-point, water cut etc).

ILI is the only m ethod that brings high confidence with re spect to e.g. inhibitor availability (and efficiency) by verifying the actual condition of the pipeline internal surface.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Types of
Tools
Free
Swimming
Cabele
Operated
Tools
Cleaning and
Batching
ToolsTools
In-Line
Inspection
Tools
Pumped
Tools
Crawler
Tools
Cleaning
Pigs
Batching
Pigs

PU-Pigs 1)Geometry
Tools
Metal Loss
Tools
Metal Loss
Tools
Batching
Pigs

Bi-Di Pigs 2)Mapping
Tools
Crack Tools In-Line
Coating
Crack Tools

Metal Loss
Tools
Speciality
Pigs
Video Pigs

Figure 4-1 Overview of different types of internal inspection and cleaning/batching tools (source: NDT Systems & Services AG).

1) PU-pigs; Polyurethane Pigs
Crack Tools
2) Bi-Di pigs; Bidirectional Pigs

Leak
Detection
Tools

4.5.5 Hydrostatic pressure testing
Pressure testing is an in dustry-accepted method for validating the in tegrity of the pip eline. The pipeline is tested up to a load of approxim ately 1.1 x opera ting pressure to 95% SMYS. The pressure test is norm ally performed as a com bined strength and leak test. Upon completion of pressure testing, the pipeline shoul d be properly cleaned, de-water ed and dried to avoid future corrosion in the system. Experience shows that local co rrosion in pipeline sys tems is of ten attributed to water leftover from the pressure test.

4.6 “Corrosion Zones” associated with external corrosion.

External surfaces of subsea pipeline systems may be divided into “corrosion zones”, based on the environmental parameters that deter mine the ac tual corrosivity. The physic al characteristics of the corrosion zones further determ ine the appli cable techn iques for co rrosion m onitoring and inspection.

The following major zones may apply;
Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Atmospheric Zone (Offshore): For pipelines on an exporting platform and on riser platform s, a “marine atmospheric zone” will apply from the pig launcher and to the uppe r limit of the riser’s splash zone. The unm itigated corrosion rate of carbon steel in a m arine atmospheric zone is in the range 0.1 to 0.3 mm/yr at ambient tem perature but m ay becom e significantly higher (0.3-1 mm/yr) for hot surfaces directly exposed to sea spray. In areas sheltered from sea spray, the unmitigated corrosion rate will approach the lower limit of the range (i.e. 0.1 mm/yr).

Closed compartments with humidity control are referred to as a “dry atmospheric zone”.

Corrosion protection means: Coating and corrosion allowance (CA) Inspection method: Visual inspection and in-line inspection (pigging)

Marine Splash Zone (Offshore): The marine splash zone can be defined as the area of e.g. a riser that is periodically in and out of the water by the influence of waves and tides. The actual length of the “splash zone” varies based on the actu al geographical location and, m oreover, from one splash zone def inition to another. Extremely corrosive conditions may prevail f or hot ris ers just above the water level; 3-10 mm /yr has been reported for risers with a wall tem perature of about 100 oC. In the splash zone below the water level, the unmitigated corrosion rate is close to that in the seawater submerged zone (see below).

Corrosion protection means: Upper zone: Coating and corrosion allowance (CA)
Lower zone: : Coating, corrosion allowance
(CA) and CP-system

Inspection method: Upper zone: Visual inspection and in-line inspection (pigging)
Lower zone: Visual inspection, in-line
inspection (pigging) and CP-monitoring

Seawater Submerged Zone; The corrosivity of the seawater submerged zone is relatively low and unmitigated carbon -steel corrosion rates in ex cess of 0.1 mm/yr would only be expected for surfaces heated by an internal fluid. For ambient temperature surfaces, the unmitigated corrosion rate is below 0.1 mm/yr, although slightly higher values may apply for local pitting attacks.

Corrosion protection means: Coating, and CP-system

Inspection method: In-line inspection (pigging) and CP- monitoring

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Offshore Buried Zone; In general, the corrosion rate in m arine sedim ents is very low (<< 0.1 mm /yr) but high local corrosion rates (o f the order of 1 mm /yr) m ay apply in the uppermost sediments if the bacterial activity is high. Internal heati ng of the pipe surface to 20 to 50 oC increases bacterial activity and hence th e potential for m icrobiologically induced corrosion. (at higher tem peratures, bacterial corrosion is inhibited). Pipeline sections covered by rock dumping, gravel and by other means are normally defined as “buried”.

Corrosion protection means: Coating, and CP-system

Inspection method: In-line inspection (pigging)

4.7 Technology Status
The available technology for m onitoring and inspection of pipelines have increased com pared to the 70’ies and pipelines that were previously difficult or even impossible to inspect, may now be accessible. The accu racy of equipment (as MF L, UT, product m easuring devises etc) has also been improved over the last years which brings a higher degree of confidence on the m onitored data and, with that, a m ore reliable assessm ent of the pipeline condition (i.e. integrity assessment). An accuracy of ±10% (at 80% con fidence) and ±0.5 mm ( at 90% confidence) is typically reported for MFL and UT pigs, respectively.

The definition of “un-piggable” lines is given as lines that are not designed for allowing standard inspection tools to pass through, which basically requi res a more or less constant bore, sufficient long radius bends and traps to launch and receiv ed pigs. Today, the inspection equip ment can be tailor-made in order to overcome the situation that was previously considered “un-piggable” with respect to standard tools.

Further, since some incidents in the past can be related to lack of industry experience (as for the 13Cr stainless steel, HISC inciden ts associated primarily with the CP-system ), th e inc reased experience f ollowing su ch events will no rmally d ecrease th e likelihood for sim ilar events to occur again.

Advanced pipeline repair and rehabilitations pr oducts and services (cove ring steel, p lastics and epoxy composite), as the Pipeline Repair System (PRS) /12/, have been developed, together with more sophisticated and robust pro grams for analysis and assessm ent of pipeline condition, allowing repair and further use of damaged pipelines.

More detailed requirements to managing system integrity are given in industry standards as API 1160, ASMEB31.8S and DNV-OS-F101.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

4.8 References
/1/ PARLOC 2001 “The update of loss of c ontainment data for offshore pipelines”, M.MacDonald Ltd for The Health and Safe ty Executive, The UK Offshore Operato rs Association and the Institute of Petroleum, July 2003
/2/ “Improving the Safety of Marine Pipe lines”, Comm ittee on the Saf ety of Marine Pipelines (USA), 1994
/3/ “Pipeline dam ages – Da mages and Incidents from Pet roleum Safety Authority’s CODAM database”, PSA Norway, August 2006
/4/ “Nedbryting av rørledninger over tid”, Rev.02, Det Kongelige Olje- og Energidepartement, 03.09.1998
/5/ DNV RP-F102 “Pipeline Field Joint Coating and Field Repair of Linepipe Coating”, October 2003
/6/ DNV RP-F106 “Factory Applied External Pipeline Coatings for Corrosion Control”, October 2003
/7/ DNV RP-F103 ”Cathodic Protection of Su bmarine Pipelines by Galvanic Anodes”, October 2003
/8/ ISO 15589-2 ”Petroleum and Natural Gas Indu stries – Cathodic Protection of Pipeline Transportation Systems; Part 2; Offshore Pipelines”, 2004 /9/ API Standard 1160, “Managing System Integrity for Hazardous Liq uid Pipelin es”, 2001
/10/ ASME B31.8S, “Managing System Integrity of Gas Pipelines”, 2001 /11/ DNV OS-F101, “Submarine Pipeline System”, 2000
/12/ DNV report No.2005-3394 “Joining Meth ods-Technology Su mmaries”, Rev.01,

12.10.2005

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
5 SUBSEA EQUIPMENT

5.1 Introduction
Subsea equipm ent will in this s ection be tem plate, m anifold and subse a Christm as tree (XT) system. The Subsea wellhead system is covered in Section 6, Drilling and wells.

5.2 Template
Templates are norm ally manufactured of low al loy carbon steel and protect ed against corrosion by a CP system . For the Norwegian continen tal sh elf, the tem plates ar e integ rated units with manifold and external fishing gear protection for protection of wellhead system s and the manifold.

The tem plate is exposed to forces that develop between the we llhead foundation, wellhead thermal growth/well temperature extension and pipeline forces. The tem plate is also seeing the gravity forces from the manifold, which the template serves as foundation for. Forces generated from wellhead growth and pipeline can over the time of operation be changing, but definitely not in a num ber of cycles that cau ses fatigue to b e a problem . However, forces experienced from trawl gear m ight have severe e ffect of the function to template structure. Due to continuous development of fishing gear, both with respect to shape and mass, can effect of those be outside original des ign criter ia of the tem plate. Detail to tem plate design such as handles, hatches, indicators etc. can be a potential snag point lead ing to sudden impact loads that might give local defects.

5.2.1 Workmanship
The tem plates are weld ed struc tures, norm ally m anufactured in low alloy carbon steel. The template is being exposed to a se t of different load scenarios th at might have the extrem e load condition during transportation/in stallation compared to in service condition. The building quality of the tem plate will d epend on the q uality of th e ste el, welds, coating, elec trical continuity straps and, if depth dictates, punctu red members. They are all falling in u nder quality assurance that can be part of manufacturing procedures and Factory Acceptance Test (FAT). The overall quality and ability to sustain any future condition of changed loads is up to what quality the template is designed for. The latter elem ent, i.e. all future load conditions, m ight be hard to detect upfront.

5.2.2 Degradation mechanism
The template, as a low alloy carbon steel stru cture shall be protecte d against corrosion through cathodic protection assisted with high quality coating. Normally epoxy based paint according to NORSOK is used. The tem plate design also ofte n includes anodes to protect wellhead and the current drainage throug h the well shall be acco unted for in CP design. Failure m echanism with respect to corrosion would therefore be related to:
1) Lack of electric continuity to all protected parts.
2) Coating not intact leading to excessive anode consumption. 3) Inadequate anode design and/or quality; anodes may fall off after some time in operation etc.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Degradation mechanism to the structural integrity to the template can be related to unfavourable effect of load combinations and operation outside original design criteria. This can be explained by change in foundation due to s ea bed erosion, new tie-ins or subsequent rock dumping outside what original designed for. Seve re global failure of a tem plate structure would be by external applied forces which can be: BOP impact loads, conductor string installation or external forces created by impact with equipm ent from non oil and gas production equipment, such as anchors and fishing gear.

5.2.3 Inspection programme
Inspection programme should typically cover following items:
• The CP system – looking for excessive consumption of anode mass and damaged/missing anodes.
• Recording of anode potential and steel if practically possible. • Coating damages, both due to general degradation and as effect of contact with fishing gear.
• General damage to structure from fishing gear.
• Hatches, handles and other elements that serve a function or can generate a snag point.
• Inspect earth cabling used to ensure electrical continuity.

5.3 Manifold
For the Norwegian sector of the North Sea the m anifold is an integral part of the template and it can be a retrievable unit from the tem plate. The manifolds today are often m anufactured from prefabricated pipes and com ponents in 22Cr duplex m aterials. Acco rding to pressure requirements and requirem ents to flexible pipe design elements can als o be m anufactured of 25Cr Super Duplex m aterials. The m echanical behaviour and m echanism of these m aterials are as described in Section 2.1.2.2 External corrosion (i.e. HISC).

After incidents, both on the No rwegian continental sh elf as other places, failure in com ponents of this m aterial have led to a number of activ ities in order to get an understanding of failure mechanism as well as developing design guidelin es (DNV RP F-112 /2/) to avoid this failure in the future. Current recommendation of usage of 22Cr Duplex and 25Cr Super Duplex m aterials have today a m ore conservative approach with respect to allowable working stress level in th e material. The design guidelines toda y includes routines and checks to allowable stress level that is not covered by former industry practice to pipe stress and component calculations.

5.3.1 Workmanship
The m anifold is a safety critical elem ent a nd involves a num ber of mechanical com ponents (valves, connectors, and sensors) that can fail. T hese components and the m anifold piping itsel f are m anufactured from advanced m aterials. T he m anifold piping are nor mally with welded connections and the valve bodies are welded in the manifold piping. Sensors tend to be bolted by approved flange styles. Welding is a critical process both with re spect to onsite p ressure testing as well as in service condition. W elds gives changes to material properties and residual stresses

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
are inher ently pre sent a t welds (po st weld hea t tre atment will reduce the r esidual str esses, if carried out). W orkmanship and co ntrol here is important both to welding qualification and non destructive testing, and the com ponents should be designed to cater for latest developm ent in design rules for HISC sensitive materials. Th e overall quality plans for m anufacturing and testing m ust follow good practice, and testing shall be conducted acco rding to the correct sequences. It is of importance that industry practice of pressure test of components and welds are done prior to coating. Piping insulation shall be according to verified and accepted procedures as well as being quality checked against these.

5.3.2 Degradation mechanism
Degradation m echanism for the m anifold can be summarized to m aterial aspects as well as functional aspects. The m anifold pi ping is m anufactured of different materials with different corrosion resistan ce. Typical m aterial selection can be Duplex (22Cr ferritic-auste nitic stee l), Super Duplex (25Cr ferritic-austenitic steel), 6M o for process bore piping, whilst control system piping can be of the less noble 316 (austenitic steel).

With respect to the external environment, all piping will be subjected to CP, not only for its own protection, but also to protect other components as valves, actuators, connectors and instruments.

Also the latter com ponents can be of different m aterials, and shoul d in general through engineering processes, be made of materials that have robustness to HISC.

If the process piping is insulated to reduce cooling effect from ambient s eawater, the insulation material shall be suf ficiently resistant to deg radation when submerged in seawater. Degradation can be a combination of absorbed seawater, w ith elevated tem perature due to hot produced fluids. Degr adation of insulation ca n also be mechanic al damage. This will be f rom operation and contact with foreign objects, or it can be due to deflection and strain absorbed by pipe material which causes the insulation material to crack. Due care shall be paid so unwanted effect of da mage to coating in combination with re duced CP. This condition can lead to accelerated degradation to the m etallic pipe m aterial when s eawater is exposed to ba re material and the CP potential is not sufficient.

Valves, valve actuators and instrum ents are a ll com ponents that will be m anufactured from different materials, where major bodies and shells can be made from low alloy carbon steel. It is therefore of im portance that these com ponents ar e electrically connected so CP protection is ensured. When degradation is discussed, this wi ll also involve m oving components. Especially override mechanisms and sliding stem s can loose so me of its intended purpose due to effects of degradation to m aterial, m arine growth, and cal careous d eposits. Th is can lead to increased friction and wear and subsequent leakages to sea from e.g. actuator housings.

Other com ponents that are of importance are clam ps and connectors. Careful design with suitable material selection should normally avoid potential problems.

With respect to the in ternal environment, internal corrosion and erosion are consid ered the most relevant failure mechanisms.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

5.3.3 Inspection programme
Manifold inspection shall cover following checks:

• General condition to pipe coating
• General condition to components, valves, actuators and instruments • Inspect for leakages at valves and components
• Inspect all connector to trees and pipelines
• Pressure test
• Visual inspection to detect foreign objects

5.4 Subsea Christmas tree (XT)
Subsea XT components are generally manufactured from low alloy carbon steel. Wetted surfaces to produced fluids are normally with corrosion resistant alloy (Inconel 625 etc). The tree itself is a construction with a number of pressure containing components bolted together with flange type connections. It is of im portance that these com ponents are assembled together af ter formalized procedures to safeguard quality.

The externals of the tree are pro tected against corrosion by a CP syste m together with a coating, normally an epoxy based coating. It is of i mportance that the m etallic material which is used is not exceeding strength and hardness requirements for the industry. Also, as mentioned, since the tree consists of several com ponents bolted togeth er it is of im portance that all components are electrically connected to ensure cathodic protection.

Internal corrosion is as m entioned in Section 2.1.2.3 and should within nor mal use not be of concern. H owever, the internals of the tree can be exposed to well stim ulation chem icals, especially after interventions that can be of aggressive nature, even when exposed to the CRA.

Some barriers may contain non metallic seals. It is reasonable to believe that all such seals could not be documented at time of design to actual working design life.

5.4.1 Workmanship
The subsea XT are a critical safety barrier as well as it is probably th e most advanced product that is permanently installed. The X Ts are built of advanced com ponents and there is a m ixture of low alloy steel com ponents as well as high a lloy stainless steel m aterials. Common for both are th at they are s ensitive to therm al effects fr om welding so it is im portant that a ll h eat treatment and welding activities are done af ter qualified and approved procedures. In contradiction to the manifold all components on a XT are connected by bolted connections, or by a flange type connection. Com pared to a weld ed pressure container there is no for mal certification of how to assem ble a bolted connecti on. It is therefore of importance that a proper good bolt torque procedures exist and are unders tood by the personnel that is doing the job. The XT are during its FAT extens ively tes ted, but l ong term effect to th e assem bly will no t be discovered, either in qualification test or in F AT. I.e. there is a discussion in the industry to whether the tree is experiencing vibration. Howeve r, regardless of vibration or its potential frequency, it is important that slim members, such as small bore piping are supported sufficiently such that long term effect of vibration does not lead to breakage.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

5.4.2 Degradation mechanism
Degradation of the tree is dependant to what it is subjected to in serv ice. Normal service should be accounted for in detail design, both with respect to exposure to produced fluids, injection chemicals as well as external e xposure to seawater with its param eters. Such parameters can be external temperature, location, and seawater depth. Degradation can also be divided into material degradation and functionality to components.

For the tree, the major building blocks are made from low alloy carbon steel with relatively high strength. Those com ponents are in most cases coated by an epoxy based coating. For som e projects the sensitivity to flow assurance is based on strict requirement to temperature insulation.

If this is the case the entire tree block should be insulated, and not only lim ited to the external process piping. Degradation of coating or insu lation can lead to reduced availability in production. High tem perature to ep oxy coated surf aces can lead to ex cessive deg radation of coating. This shall however be accounted fo r in the system CP design. For tem perature insulation-material care shall be tak en to seawat er absorption, mechanical wear fro m impact of external components, a nd loss of binding betw een insulation and steel m aterial. The latter ca n lead to severe local corrosion if the com bination is unfavourable wrt seawater access and lack of CP effect.

High strength m aterial, both of carb on steel as well as sophisticated stainless qualities, shall be selected with sufficient m argin to avoid HISC, and the operati on stress level shall be below certain values.

Valves, actuators, conn ectors a nd instrum ents are m anufactured from a m ixture of different metallic materials. The large shells and bodies are usually manufactured from low alloy carbon steel, and sm aller components are likely to have some stainless type m aterial, e.g. Duplex and Super Duplex. Override m echanisms and override stem s that extend from actuator housing to external seawater atmosphere can be a potential degradation to functionality. E.g. marine growth, calcareous d eposits can lead to failu re of th e sealing elem ent between actuators an d seawater allowing se awater in to areas no t to lerant to th is. Also severe f riction can, due to m entioned effects, lead to problems in operating e.g. valves or other mechanisms.

The internals of the tree shall in normal service not be of concern. Through material selection the tree shall be robust. However, it seem s to be s ituations w ere the well has been treated with chemicals than can h ave severe effect to tree internals when not flushe d out satisfactory. These chemicals (which can be acid) can rem ain in d ead end pockets, in seal grooves etc. and cause severe local damage. Effect to soft seals can be unknown.

5.4.3 Inspection program
Inspection of Christmas tree shall cover following checks:
• The CP system – looking for excessive consumption of anode mass • Recording of anode potential and steel if practically • Coating damages, both due to general degradation and as effect of hot surfaces • General damage to structure from fishing gear
• General condition to pipe coating
• Inspect for leakages at valves, connectors, sensors and components • Pressure test
• Visual inspection to detect foreign objects

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

5.5 References
/1/ DNV-RP-F112 Design of duplex stainless st eel subsea equipm ent exposed to cathodic protection, Draft issue April 2006

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
6 DRILLING AND WELLS

6.1 Introduction
As drilling equipment ages, the operator has several new challenges to consider, such as:

– Changes in integrity, e.g. time dependent modes such as corrosion and fatigue, or random modes such as mechanical wear.
– Accidental events (incl. well settlements)
– Requirements to operate beyond the design lifetime.
– Design and Approval no longer valid due to the above mentioned issues.

6.2 History
Generally, review and analysis of historically causes of drilling equipm ent failures worldwide, indicate that mechanical wear, is the m ost widely reported c ause of failure for offshore drilling equipment and wells, followed by corrosion and fatigue damage.

6.3 Main degradation mechanisms
Threats shall be systematica lly identified, assessed and doc umented throughout the operational lifetime. This shall be d one for each com ponent and for the system . Examples of typical th reats are:

— mechanical wear
— corrosion
— Erosion
— buckling
— fatigue

6.4 Drilling system and well system
A typical schematic layout of the drilling equipment is shown in Figure 6-1.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 6-1 Typical schematic layout, drilling equipment (Aker Kværner).

Crown block
Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

6.4.1 Pipe handling
The pipe handling system often experiences m echanical wear due to fr equent use. Com ponents experiencing such wear should be inspected pe riodically. Furtherm ore, im pacts and vibration during pipe handling m ight lead to loss of pret ension in bolted connections. Loss of pretension increases the probability for fatigue dam age i n the connections. The pretension of the bolted connections should be checked periodically during the design life.

6.4.2 Main hoisting system
The brake capacity of the brakes o n the draw work might get reduced due to oil c ontamination on the discs. An important aspect of draw work brakes is the friction between the brake pads and the discs. The friction factor used in the brak e design calculations is to DNV’s experience often non-conservative.

The main brake is an electric m otor, and th e driller training using the brak e system is often not satisfactory, which lead s to po tential hazardous situations. This also includes th at some drilling rigs uses the emergency brake on a daily basis in the drilling operation, which reduces the safety level in the main hoisting system.

If the pads are not “run-in” properly, a reduced brake capacity will be experienced. This may not be noticed during norm al operations, but m ay lead to an accid ent in an em ergency brak ing operation. The disc springs of the callipers are s ubjected to low-cycle f atigue which reduces the brake capacity after 1-2 years of operation. The fabrication tolerances of the callipers are sometimes too inaccu rate, which g ive a h igher airgap for th e disc brak es and follo wing loss of brake effect on one of the sides.

The travelling block/crown blocks are often experien cing m echanical wear in sh eave groov es during their life time.

The safety level of the draw work and the crow n block should be at the sam e level. The crown block safety level is less than the draw wor k, which has lead to hazardous situations during accidental events where details in the crown block have collapsed.

In general, the development of the drilling operations goes toward higher hook loads, resulting in a need for corresponding increased brake capacity.

6.4.3 Control systems
Experience shows that Control system s som etimes contain programm ing errors, leading to logical errors in the system and planned operations fail.

After a few years in operation, whole or parts of a control system might be upgraded. The quality control of the new control system is of ten not as exte nsive as when the drilling s ystem was initially designed, m anufactured and tested. Experience shows that such upgrading som etimes lead to unwanted events. Safety assessm ents when upgrading control systems should be increased to prevent uncontrolle d situations/operations. In th e draw work, several control systems from different vendors are often not suffici ently correlated. This aspect gives a source for failure during operation. To prevent this so urce for failure, one of the vendors should be given the main responsibility with respect to total quality assurance, documentation and interface aspects.

6.4.4 Iron Roughneck and pipe racking system
The iron roughneck as well as the pipe racking sy stem experiences extensive m echanical wear, and are often replaced/upgraded after 4-5 years.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

6.4.5 Steel wire ropes
Steel wire ropes are subject to continuous wear and fatigue loading. On drawwork and riser tension system there is implemented a ton*mileage measuring device. After a specific ton*mileage the cut and slip is carried out. It is normal procedure to send part of the used wire to a test laboratory in order to calibrate the acceptance level for cut& slip. Worn steel wire ropes may be reduced in diameter due to roll out and fatigue cracks may be introduced.

If steel wire rope are subject to high temperature from for example flaring the wire grease may be lost which lead to more rapid degradation of the wire than expected.

Damage to wire, caused by kinking, running over sharp edges or due to bad spooling on winches are common causes of replacement for wires in winches.

Winches which lack spooling devices m ay e xperience incorrect spooli ng. This effect gives uneven contact on the wire and leads to higher wear on the wire.

6.4.6 Sealed machinery
In general, we have experienced that closed sealed machinery sometimes are not properly sealed, and contamination is able to enter an d cause extensive mechanical wear on the machinery. This is hidden errors which are not found during extern al visual inspection. Typical examples: splines and bearings.

6.4.7 Drill string
The drill string often experience fatigue cracking due to ageing and poor control with the number of load cycles. The handling and use of the dr ill string give sm all dam ages which lead to corrosion, which accelerates fatigu e (i. e. corro sion fatigu e as describ ed in Section 2.1.4) and thereby give a fatigue weakness and dam age. High differential pressures combined with eroding environment might lead to wash-out of the drill string as shown in Figure 6-2.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 6-2 Typical example wash-out of drill string (DNV photo).

6.4.8 Drilling riser
The drilling riser is often desi gned with a sm aller thickness amendment due to co rrosion than other risers and structures, but should be s ubject to a higher control regim e. Reduced wall thickness due to corrosion and er osion leads to a reduced tens ion capacity. Through the riser management system on the rig, there should be a system to rotate the position of the ris er elements periodically to be in the sp lash zone as well as in the highest loaded positions in turn.

There should also be a load record of the riser stack giving an overview of the loading of eac h component.

6.5 Blow Out Preventer (BOP)

The gaskets of the BOP are often subject to a continuous mechanical wear during the drilling operation as well as periodic testing of the BOP. In some kinds of BOPs, the shear ram can wear out after only approx. 15 runs. In such cases, it is of vital importance that the operator has exact control of the his tory of the BOP, and that th e critical gaskets are repl aced before leaks o ccur.

Degradation and dam age of som e of the BOP ga skets can increase significantly due to special operations such as drilling through casing where st eel swarf from the dril ling operation passes Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
the BOP. The feasibility of the gaskets is periodi cally controlled by pressure tests demonstrating that the functionality of the BOP is maintained.

6.6 Subsea Wellhead
The Subsea wellhead is norm ally m anufactured fr om relatively high strength low alloy CMn steel. Sealing surfaces are normally inlay welded with CRA. The wellhead are welded to a casing pup piece, often m anufactured by API 5L grade m aterials. Current analysis shows that the intended design and m echanical behaviour are dependant of the level and quality of cement fill between 20” casing and 30” conduc tor. The different mechanical behaviour is of particular importance for the wellheads cap acity to take riser fatigue loads. Current de sign load as defined by NORSOK U-001 are not sufficient to describe the capacity to a wellh ead system. Recent work by DNV also indicates that the wellhead doe s not provide the sam e level of conservatism as the common industry standard for completion/work over risers.

The potential dam age to Subsea wellheads are re lated to drilling or workover mode when the wellheads are subjected to a rise r load. The f ailure mode is rela ted to f atigue, both in welds as well as at stress risers in base m aterial. As described in Section 2.1.3 in this docum ent, a fatigue failure will have an initiation s tage with slo w propagation before reaching a lim it where remaining cross section is overloaded and a ra pid failure develops. The wellheads are norm ally not accessible for inspection, and hence it is difficult to detect cracks that are in initiation stage.

Recent assessments of fati gue lifetime show that the wellh eads are u tilised at a lev el exceeding already used tim e with riser exposure. This is particularly im portant when assessm ents of old wellhead systems are d one with respect to IO R programmes which m ay lead to ex tended riser exposure.

6.6.1 Workmanship
The Subsea wellheads are m anufactured of relatively high strength CMn steel. As the wellheads can show short fatigue life, it is important th at th ey are designed with prof iles that giv e lo w stress concentration factors. All welding and heat treatment must be done according to approved procedures and subsequently inspected by N DT. The fa tigue resistance of welds is very dependant of the weld configuration. A weld that is grinded after welding will have a better fatigue life com pared to a non m achined weld. Ho wever, it is of im portance that the weld between 18 ¾” wellhead housing to 20” casing weld gives lim itation to fi nal surface treatm ent after weldin g. This is due to restricted access for personnel, due to lo ng assem bly and s mall internal diameter.

The cement job in the a nnuli between 20” and 30” casings is dif ficult to check. Cem ent will be contaminated as well as it is of low strength type with low viscosity. During curing it will shr ink and give less support. T hermal expansion due to temperature variations w ill also decrease the supporting effect from the cement. This can be unf avourable with respect to the fatigue life of the 20” casing and its welds.

6.7 Conductors
The conductors are norm ally m anufactured from low alloy carbon steel and are dependant of cathodic protection against extern al corrosion. As m entioned above for Subsea wellheads, the conductors are also welded to nom inal thickne ss API 5 L grade steel m aterials. This weld together with stress ris ers from changes in cros s sectional area are all elem ents that can giv e

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
reduced f atigue lif e c apacity to th e wellhe ad system. Fatigue loads are experienced by the conductor when it is subjected to completion/workover risers- as well as to a drilling riser set up.

The conductor capacity is dependan t of quality and support externally provided by g routing and strength capacity to surrounding sea bed. The c onductor capacity is also influenced by lateral support from a te mplate system . In other words, a satellite conductor w ithout support from a template structure is exposed to higher external loads. This leads to lower static capacity an d shorter fatigue life.

As mentioned above for Subsea wellheads, the c onductors’ mechanical behaviour are not always as expected through design. It is a com bined unit with wellhead a nd they are having a mechanical interaction that de pends on factors m entioned above, such as external grouting and internal cement quality in the annuli between 20” and 30”.

For combinations of the factors above, the wellhead/conductor capacities to riser fatigue load can be relatively small. In some cases they are calculated to as low as shorter than a normal duration of an intervention cam paign. Theref ore the we llhead and conductor show lower conservatis m than what is acceptable for a riser set-up.

It is im portant to include such evaluations into the process when IOR programm es or other programmes are discussed that lead to extended riser exposure to existing wellhead system.

The conductors are easier to in spect com pared to the wellh ead housing. However, the m ost severe failure m ode, fatigue m ight be im possible to inspect for. Also du e to m ost critical weld might be in an area where external inspection is impossible due to external frames, and internally due to all casings that is installed.

6.7.1 Workmanship
The conductor is constructed of m ild steel. It is not pressure containing so it is only seeing mechanical load. Robustness to fatigue life is very dependant on the quality of the weld.

Compared to wellhead housings , the weld betw een conductor housing and conductor casing is possible to both grind and inspect. However, it is not certain that this have been accounted for at the time of design and m anufacture to old system s. External grouting and possible fracturing of the seabed is factors that can lead to reduced capacity to both static bending loads as well as fatigue load. A tem plate installe d conductor is norm ally better in those respects than satellite wells.

6.8 Production casing in wells
Depending on the well behaviour, the Production casing and other components in the completion string will be subject to corro sion and erosion. F or wells with a high degree of sand production, erosion (see Section 2.1 .1) will b e a relevant degradation mechanism. For wells sub ject to gas lift, frequent start/stop, water production or varying injection of water/gas, corrosion will be th e main degradation m echanism. Geotechnical s cenarios lik e settlem ents, dislocations, etc., can introduce addition al sh ear and co mpression loads, and s hould be taken into account in th e design. Detection of damaged production casing is primarily performed by pressure surveillance, and there are a num ber of quality control and surveillance methods of production casing with respect to thickness measuring.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

6.9 High pressure tubing and manifolds for circulation of drill mud In high pressure systems, erosion will be the main degradation mechanism. Especially in choke- and kill manifolds during well control situ ations, very large erosion rate s will appear due to the mixture of gas and drilling slurry. Also in hi gh-pressure piping handling drill m ud and cem ent, erosion is the main degradation mechanism, especially in bends and branches.

6.10 Recertification of well control equipment
The conten t of this section is mainly extracted fr om ref . /1/. It is DNV’s interpretation of the PSA regulations that a m ajor overhaul/inspection with verification of Blow Out Preventers and other pressure control equipment used for Drilling, Completion and Workover operations, should be performed at least every five years.

In addition, the need to recertify equipment can occur due to several other causes:

- Change of intended use / loading aspects
- Increasing original design life
- Repair of equipment

The purpos e of this in spection is to ver ify an d docum ent that the eq uipment condition and properties are within the original acceptance criteria.

The extent of inspection may be influenced on the following parameters:
• Repair history
• Maintenance history
• Operational history
• Manufacturers guidelines
• Change in rules and regulations or company’s governing documents

The following activities shall be included in the recertification process:
• Review of original documentation with special focus on traceability. • Review of m aintenance history/records, to verify the am ount of use and extent of maintenance
• Stripping/dismantling of equipment
• Visual inspection.
• NDT
• Dimensional check of selected components/review of dimensional check reports. • Change out of seals, treads etc.
• Reassembly – recoating - preservation
• Load/pressure testing and functional testing.

The acceptance criteria for the various ins pections perform ed shall be b ased on the manufacturer’s initia l qualification program s and engineer ing docum entation, as well as internationally recognized codes and standards.

DNV does not recommend recertifying equipment unless the acceptance criteria applied gives a certain conf idence with regards to m argins to f ailure. It must be possible to verify that the

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
functional, performance and safety m argins of the equipment are with in the origin al acceptance criteria.

6.11 References
/1/ DNV-OTG-06 Recertification of well control equipment – service de scription, Septem ber 2005

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
7 MOORING SYSTEM

7.1 Introduction
A Mooring Integrity JIP carried out by Noble Denton Europe /1/ has concluded as follows:
- The interface between the surface vessel and the mooring line requires particular attention for all types of FPS.
- Carefully planned innovative 1 inspection, m aking use of a ll possible tools, has been demonstrated to b e able to d etect problems relatively early before they become a potential source of failure.
- At present no in-water techniques exist to check for possible fatigue cracks. - On two North Sea FPSs chain wear and corrosion have been found to be significantly higher than what is specified by m ost mooring de sign codes. This wear seem s to be m ore pronounced on less heavily loaded leeward lines com pared to the m ore loaded w indward lines.
- At present there is little data available which indicates how the break strength of long ter m deployed mooring components will be reduced by wear, corrosion incl uding pitting and the possible development of small fatigue cracks.
- A possible contributory m echanism for the re latively high line failure rate am ong drilling semi-submersibles has been identified. This is believed to be due to rigs thinking they have set up balanced pre-tensions, when in fact this has not been achieved. One reason can be that the lo ad cells on th e windlasses are not calib rated properly. If the tension m eters are well positioned, working properly and their calibration is in date, a likely cause of unbalanced line tensions is partial seizure of the gypsy wheel s. This can be confirmed by a simple line Payout/Pull-In test. If this reveals that som e of the gypsy wheels are partially se ized, an attempt should be m ade to free them up. Howeve r, if the unit is on station it m ay not be feasible to undertake such work in situ. In such a case the lin e tensions out with the f airlead should be determined by other measures such as:
o ROV or pos sibly diver monitoring of the chain angles where they em erge from the fairleads
o Acoustic monitoring of the x, y a nd z positio ns of specific connecto rs on the mooring lines
From these measurements it is possible to back calculate the actual line tensions as long as this is done in calm conditions with minimal tidal variations.

7.2 Fairleads and chain stoppers
Typical problems with fairleads are malfunction of bearings, excessive wear and te ar in fairlead wheels and pockets due to insufficient support for chain. This m ay be caused by low tension and/or the f airlead is n ot rota ting with the chain, which m ay be caused by bearing problem s.

Excessive wear and tear has b een discovered in cas es were th e chain is term inated in chain stoppers underneath the unit.

1 The mooring systems of mobile offshore units are inspected onshore within a period of 5 years. For units permanently installed at a location the inspection has to be carried out offshore and it is important to use all possible available tools. In situ water inspection techniques are contin uing to improve, but further developments are needed to provide dimensional data on links all around the inter-grip area and to improve the marine growth cleaning off speed. For further information see /1/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

7.3 Chain
Historically, chain m anufactured in the 1980’s, es pecially Grade 4 chain suffered with quality control problems and subsequent br ittle fracture problems. Brittle fracture is the term used for a rapid failure of m aterial, which involves low ductility. It is pa rtly dependent on th e type of steel used, or the process ing that it has been subject to, and is exacerbated b y stress-raisers or cracks in the material. It can result in the f ailure of chains at rela tively low tensions. The problem with brittle fracture is that the propensity of the material to fail in this way is not obvious to the naked eye and can only be quantifie d by destructive testing. Steel is more susceptible to brittle fracture when the yield strength is high or where the op erational tem peratures are low. T he welding process, and subsequent heat treatm ent, used to form the chain link dur ing manufacture must be very ca refully con trolled to prev ent br ittle fracture p roblems with chain. This said the metallurgical and m anufacturing issues appear to have been largely res olved such that m odern high strength chain can now be consistently produced.

Much of this problem atic chain has now been removed or scrapped but the associated problems have had an im pact on the industry over the years /2/. It is also pos sible that there is a residual amount of this chain around.

The prime cause of line failure now appears to be with the connecting shackles or with links that have been mechanically damaged. Common modes of failure in chain systems therefore include:

- Mechanical damage to links
- Missing or loose studs
- Failure of connecting links
- Brittle fracture of links (not so common with improved quality control of chain) - Fatigue

Missing or loose studs has no di rectly influen ce on the b reaking strength, however the s tress distribution in the link is changed and the footpr int will rep resent an area where fatigue crack s can develop and result in fatigue failure of the link. Control with stud pre ssing is essential and instances have been seen where studs have been pressed without having been correctly seated in the imprints. Also, studs have been expanded by excessive am ounts with detrim ental effect on the links. Ideally, stud pressing s hould result in light contact with the link. There is no limitation regarding how many times a stud can be pressed.

The following tolerances regarding studs apply /8/:
- Axial stud movement up to 1 mm is acceptable.
- Axial stud movement greater than 2 mm is unacceptable. - Links are to be removed or studs are to be pressed using an approved procedure. - Acceptance of axial stud movement from 1 to 2 mm must be evaluated based on the environmental conditions of the unit’s location and expected period of time before the chain is again available for inspection.
- Lateral movement up to 4 mm is acceptable provided there is no realistic prospect of the stud falling out.
- Welding of studs is not acceptable.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

In Figure 7-1 the line failure was caused by fa tigue. The fatigue initiation has caused the chain link fracture and growth has been probably due to overloading of the chai n link. Most likely the overloading has been probably due to twisting of the chain link, in addition to high cyclic axial tension loading of the mooring chain.

Figure 7-1 Fatigue (DNV Photo).

The chain links shown in Figure 7-2 have suffere d bacterial corrosion along one of the straight sides. The corrosion rate was estim ated to be of 2.5 mm /year at the co ntact area between chain and seabed based on this exam ination. Even t hough the dam age was confined to the straight sides of the chain links (areas away from those with the h ighest potential stress es), a corro sion rate of this order can obviously affect life of a mooring system that is designed to be in operation during a 20 year period. In Figure 7-3 a crack was detected, which was initiated in the foot print area.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 7-2 Bacterial corrosion (DNV classed FPSO offshore West Africa).

Figure 7-3 Crack initiated in the stud foot print (DNV photo).

Figure 7-4 shows severe wear and tear in a long term mooring system.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 7-4 Wear and tear in chain links (Photo Norsk Hydro).

7.4 Recertification of chain
Normally recertified chain is not accepted in long term mooring. However, for drilling units it is more common to rent used chain if the unit’s own mooring system is not sufficient for a new location. Recertification of chain shall be carrie d out applying the sam e inspection requirem ent as for a complete periodical survey, which incl ude visual examination, extensive non-destructive testing, dimension control and pressing of studs.

A recertified chain for m obile offshore units shall p ass the requirements for ren ewal survey given in D NV Instruction to surve yor /8/. It is generally not possible to state that a recertified chain is as good as a new equivalent chain. However, the recertified chain is found good enough for 5 year in operation, since a ren ewal survey is required every 5 years. Recertified chain is normally not accepted for permanent installations.

7.5 Synthetic fibre ropes
Typical failure mechanisms are:

- Ingress of particles
- Mechanical damage during installation and hook up activities - Mechanical damage caused be fishing activities
- Creep

Ingress of particles such as sand or clay into the load bearin g part of the f ibre rope will reduc e the breaking strength of the fibre rope significan tly. This problem can be avoided by installing a filter underneath the outer jacket of the fibre ropes. DNV has qualified such filters for Marlow Ropes /3/ and ScanRope /4/.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT
Mechanical damage during installation, hook up a nd ROV operations must be avoided. Further, trawling has caused failure of fibre ropes. DNV has developed a recommended practice /5/ to assess the rest capacity of a damaged fibre rope.

The purpose of this recommended pr actice is to provide assessm ent basis for the suitability of a polyester mooring rope to remain in service, after it has been m echanically damaged by external objects. The recomm ended practice is applicab le to any "parallel-subrope" type of rope. The inputs required to perform the necessary calcul ations are provided by the rope m anufacturer.

This information is given in the M anufacturer's Report. The dam age assessment is based on the subrope-to-rope relationship, since the subrope is the prim ary building block of the rope.

Subrope-to-rope assessm ent is re quired since the effect of da mage is highly dependent on the damage distribution. This im plies that for a give n loss of area, th e resulting rope streng th and fatigue performance will vary depending on the distribution of the damage.

Synthetic ropes have become an accepted alternative to chain and steel wire rope mooring lines in recen t y ears. At present, polyes ter fibre is the m ost widely used synthetic fibre for this purpose. High m odulus polyethylene (HMPE) is an alternative to pol yester, with m any favourable properties. However, HMPE is m ore susceptible to creep than polyester, and this behaviour requires careful assessment as part of the design process for HMPE mooring lines.

Creep is a for m of perm anent elongation of synthetic fibres. Th e creep rate increases with increasing specific load and temperature. Creep can ultimately lead to f ailure of a mooring line.

Creep need not be a serious problem if it is properly accounted for in the design of a mooring line.

7.6 Steel wire ropes
There are three types of steel wire ro pes used in mooring systems (see Figure 7-5) with different expected service life:

- Six and multi strand, normally used by mobile offshore units. - Spiral strand with and without plastic sheathing, normally used by permanent installed units. - Half and fully locked coil wi th or without plastic sheathi ng, norm ally used by perm anent installed units.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Figure 7-5 Constructions of Steel wire ropes /6/.

Failure of wire in mooring lines is caused by one of three causes:

- Mechanical damage to the wire
- Corrosion/Wear
- Fatigue
- Chasing operations can also cause bends and kinks in mooring wires. - Bends m ay not be serio us enough to replace th e wire rope; however, k inks will seriously reduce strength.

With respect to design life the following table from /6, 7/ can be used as a rough guideline:

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Choice of steel wire rope construction
Possibilities for replacement of wire
rope segments
Field
design life
(years) Yes No
< 8 A/B/C A/B/C
8 – 15 A/B/C A/B
> 15 A/B A
A) Half locked coil/full locked coil/spiral strand
with plastic sheathing
B) Half locked coil/full locked coil/spiral strand
without plastic sheathing
C) Six strand/multi strand

A common design requirem ent is that wire rope segm ents in m ooring lines are to be protected against corrosion attacks throughout the design life. The w ire rope is therefore assum ed to be fully protected such that its fatigue life approach es that in a ir. This is no rmally ensured by the following measures or combinations thereof:

- Sacrificial coating of individual wires.
- Application of blocking com pound on each layer of th e strand du ring stranding. The compound should fill all crevices in the wire rope, strongly adhere to individual wire surfaces and have good lubrication properties.
- Cathodic protection by spinning zinc or other sacrificial anode alloy wires in one of the outer 3 layers of wire rope during manufacture.
- Surface sheathing of the wire rope by an extruded plastic jacket in order to prevent ingress of sea water and flushing out the blocking compound.

The ends of each wire rope segm ent are norm ally terminated with sockets. Free bend ing at the sockets outlet can reduce the wire rope fatigue life. To avoid prem ature fatigue failure, a bend limiting device is of ten incorporated at these locations. Such a dev ice is designed to sm oothly transfer the loads from the sockets to the rope. To prevent ingress of water in the socket a sealing system may be incorporated in the device.

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

DET NORSKE VERITAS
Report No: 2006-3496, rev. 0

TECHNICAL REPORT

Reference to part of this report which may lead to misinterpretation is not permissible.

AACAACDW.doc

7.7 References
/1/ Floating production system - JIP FPS moor ing integrity. Prepared by Noble Denton Europe Limited for the Health and Safety Executive 2006 /2/ P. J. Donaldson, B.Sc. Master Marine r, MRINA M. Brown B.Sc., M.Sc., M.B.A., C.Eng., MRINA M. Pithie, Master Mariner: “Design and integrity management of mobile installation moorings”. Noble Denton Europe Ltd, 2004 /3/ DNV Report No. BGN-R3100270 “Test of Soil Filter”, July 2000 /4/ DNV Report No. 2005-3502 “Testing of Soil Filter”
/5/ DNV-RP-E304 “Damage Assessment of Fibre Ropes for Offshore Mooring” /6/ DNV Certification Note N o.2.5 “Certification of Offshore Mooring Steel Wire Ropes, May 1995
/7/ DNV-OS-E301 “Position Mooring”, October 2004
/8/ DNV Instruction to surveyors for Classi fication of Mobile Offshore Units, I-C3.4 Mooring System

- o0o -

DET NORSKE VERITAS

APPENDIX
A

SUMMARY

Page A-1
Report No. 2006-3496, rev. 0

DET NORSKE VERITAS

SUMMARY
Area Relevant constructions, systems or equipment on installations Relevant degradation mechanisms Typical failure modes Load bearing
structures,
concrete
- All concrete on the structure including the skirt below seabed. - Chemical seawater attack - Freezing and thawing
- Expansive alkali reactions
- Bacterial corrosion
- Chloride penetration of the concrete
- Galvanic corrosion of the
reinforcement steel

- Cracking
- Fracture
Load bearing
structures, steel
- Floating offshore unit
- Jack-up rig
- Column-stabilised units
- Corrosion, fatigue

- Local structural
damage, fatigue
cracking
Subsea pipelines - Service/process conditions

- Free spans
- Coating
- Asphalt enamel coating
- Weight coating of pipeline
- Atmospheric zone, marine splash zone, seawater submerged zone offshore buried zone
- Corrosion (internal)
- Sulphide stress cracking
- Fatigue
- Third party damage (TPD)
- Spalling
- Spalling
- Corrosion (internal/external)

- Leakage
- Burst
- Local
buckling/collapse
- Reduced wall
thickness
Subsea
equipment
- Template

- Manifold

- Subsea XT system
- Third party damage (fishing gear)
- External corrosion
- External/internal corrosion
- HISC
- Mechanical damage to coating
- Erosion
- Wear
- External corrosion
- HISC
- Coating damage
- Cracking
- Reduced wall
thickness
Page A-2
Report No. 2006-3496, rev. 0

DET NORSKE VERITAS
Drilling and wells - Pipe handling system

- Main hoisting system

- Iron roughneck and pipe racking system
- Steel wire rope

- Sealed machinery
- Drill string

- Drilling riser

- BOP
- Subsea wellhead
- Conductors
- Production casing in wells

- High pressure tubing and manifolds for circulation of drill mud - Mechanical wear
- Fatigue
- Low cycle fatigue
- Mechanical wear
- Mechanical wear
- Fatigue
- Wear
- Wear
- Fatigue
- Corrosion fatigue
- Corrosion
- Erosion
- Mechanical wear
- Fatigue
- Fatigue
- Corrosion
- Erosion
- Erosion
- Fatigue cracking
- Reduced wall
thickness
Mooring system - Fairleads and chain stoppers
- Chain

- Synthetic fibre ropes

- Steel wire ropes
- Wear and tear
- Bacterial corrosion
- Mechanical damage to links
- Wear and tear
- Brittle material (welding related)
- Fatigue
- Ingress of particles
- Mechanical damage during
installation or due to fishing activity
- Creep
- Mechanical damage to wire
- Corrosion/Wear
- Fatigue
- Failure of mooring
line
- Reduced wall
thickness
- Fatigue cracking

- o0o -
Page A-3
Report No. 2006-3496, rev. 0

3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 1/8

SKIP TO MAIN CONTENT

ENERGY & ENVIRONMENT
California’s latest power grid problems are just the beginning State officials knew ahead of the recent heat wave that the grid was on shaky ground.

Residents on the Central Coast had tussled for decades over the role of Diablo Canyon, but until recently, they thought the debate was over. | Michael A. Mariant/AP Photo By CAMILLE VON KAENEL
09/23/2022 04:30 AM EDT
SACRAMENTO, Calif. — An epic heat wave this summer brought California’s power grid to the brink of collapse, and put its governor on defense as he touted the state’s nation-leading climate goals.


3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 2/8 In Democratic Gov. Gavin Newsom’s telling, the state kept the lights on because of its efforts to bolster renewable energy.
“Went right up to the edge of breaking our grid, but it didn’t,” Newsom said at a Clinton Global Initiative event this week, describing this month’s scorcher to dignitaries gathered in New York City for Climate Week at the U.N. “This transition worked.”
The reality, however, is a lot messier.

California’s recent decisions to postpone the closure of its last nuclear plant and to extend the life of some natural gas-fired facilities highlight what officials and experts say is the fact that the state with the most ambitious energy goals is far from achieving them.

Growing demand for electricity and the fickle nature, for now, of greener technologies such as wind and solar are making it hard to progress toward the state-mandated goal of a grid that’s 100 percent emissions-free by 2045.

Renewables provided 36 percent of the state’s power supply on average so far this year.

Those constraints were behind the recent decision by the Legislature, at Newsom’s urging, to postpone the retirement of the Diablo Canyon nuclear plant despite the fact that activists thought they’d secured its closure — and the governor himself once supported the idea.
3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 3/8 Those constraints were behind the recent decision by the Legislature, at Gavin Newsom’s urging, to postpone the retirement of the Diablo Canyon nuclear plant despite the fact that activists thought they'd secured its closure — and the governor himself once supported the idea. | Michael A. Mariant/AP Photo The 10 days of triple-digit temperatures across the state this month sent power demand surging to a record level, bringing state regulators close to ordering rolling blackouts, a potentially deadly move and a political disaster.

It was the realization of a nightmare scenario a top state energy official said he’s been considering for months.
“Oh, my lord, we are in a very bad situation compared to even the worst case that we anticipated,” Siva Gunda, vice chair of the California Energy Commission, said he recalls thinking in the spring, when supply chain delays and a tariff on solar imports — compounded by severe drought — started to look like a multi-year power crisis.

The possibility of rolling blackouts became a shadow looming over California Democrats, even those who felt uneasy about keeping Diablo Canyon open.

Some talked publicly about how outages contributed to the impeachment of then-Gov. Gray Davis at the beginning of the century.

3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 4/8 What’s needed now, officials say, is even more investment by the state akin to the Marshall Plan that rebuilt Europe after World War II. “Enough isn’t being done right now” to avoid a worrying gap in the power supply in the future, said state Sen. John Laird, a Democrat from Santa Cruz who has argued the state needs massive new investment in renewable energy and batteries to move off fossil fuels.
“We have to make sure that we have more wind, we have more solar, we really develop offshore wind, get out of the way of some of the developed renewables so that they come on the grid,” Laird said.

The stakes are existential. After the Enron electricity trading scandal and the Western energy crisis that followed 20 years ago, the state’s reputation was in tatters and a governor got recalled. But California quickly built itself up as a model for embracing wind and solar power — becoming the measure against which other states compare their own climate ambitions.

Now, by a state estimate, California will need to deploy renewable energy at five times its average pace to meet its mandated goal of 100 percent emissions- free power by 2045. All that while contending with rising temperatures, drought and wildfire.

That uncomfortable reality gave Laird some sleepless nights, he recalled in an interview. He is close friends with people who fought for decades to shut the nuclear plant, worried there might be an accident along the seismically active Central Coast, among other concerns.

He eventually voted to keep Diablo Canyon open, a difficult decision he said was driven by projections that California would not have enough new wind and solar power in time to make up for its closure. The last-minute scramble by Newsom and the Legislature could postpone the plant’s demise until 2030, reversing a deal made six years ago between green groups, labor and regulators to close it in 2025. The nuclear plant provides up to ten percent of the state’s power.
3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 5/8 Laird sees the delay, which requires federal approval, as a stopgap measure that shouldn’t get in the way of a massive build-up of renewable energy. “We need it either way,” he said about the infusion of renewable power. “If Diablo is extended, we need it. If it’s not extended, we need it. And one of the reasons we’re here is that not enough was brought online.” The state senator’s district is a microcosm of California’s energy transition.

Criss-crossed by high-voltage transmission lines, the region not only includes the heavily fortified Diablo Canyon, nestled out of sight among rolling green hills along the ocean, but also the first proposed offshore wind farm on the West Coast. Projects like this are meant to help the state get rid of all carbon emissions — if they can be built on time.

Workers along the Central Coast have a long history of building energy projects and see the state’s grid challenges as an opportunity, Dawn Ortiz-Legg, a San Luis Obispo County supervisor, said in an interview in Morro Bay, the sleepy fishing village slated to host the floating offshore wind turbines 20 miles off its shore. The county official previously helped turn homebuilders into workers who constructed large solar farms that helped lift the county out of recession.

Behind Ortiz-Legg rose three fog-shrouded, iconic smokestacks at a shuttered power plant that once burned coal and gas. Now, in an apt metaphor, they are scheduled to come down to make way for a proposed battery facility that would store renewable power.

Ortiz-Legg said she agreed with postponing the closure of the nuclear power plant — but only temporarily. Now, she’s calling for state leaders to act with more urgency to make sure Diablo Canyon can be closed, according to the new plan, in 2030.
“It’s really important to note that in 2001, 2002, when California had its energy crisis, gas plants were permitted in 20 days,” she said.

The tight timeline is making Jane Swanson skeptical. A nearby resident, she has fought the nuclear plant since the 1970s out of concerns about nuclear 3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 6/8 waste and seismic safety, as part of the organization Mothers for Peace. She was looking forward to Pacific Gas & Electric, which runs the plant, closing it down and now fears the future will bring more broken promises. “You can’t believe a word PG&E says,” Swanson said. “You can’t believe a word politicians say. So I have no faith that any future agreement or end date will be observed because we’ve had those dates and they’re not being observed.” State energy regulators last year ordered utilities to add more clean energy to the grid over the next three years than ever before. But getting the solar panels and wind turbines built and plugged in depends largely on the whims of the global economy.

California energy officials went into crisis mode this spring when supply chain issues from the pandemic and a new retroactive tariff on solar panels cast a pall over the industry, delaying the renewable energy projects they had been counting on before Diablo Canyon was set to retire.

No one disputes that an additional margin of safety is needed, said Ralph Cavanagh, the energy co-director for the Natural Resources Defense Council.

Still, he argued, the governor locked down on Diablo Canyon prematurely instead of looking at alternatives like bolstering energy efficiency and pulling more energy from other Western states.

Republicans in California, meanwhile, prefer the nuclear power option. “If the plant gets decommissioned, we don’t have enough juice to keep the lights on and keep air conditioners working and keep people’s EVs charged,” said Assemblymember Jordan Cunningham, the Republican who represents the Central Coast and thinks the plant should keep running even longer.

Cunningham also largely supports California’s climate goals and helped shape the ambitious offshore wind target for the state.
“I think in five years’ time, we’ll be in a better place, with renewables coupled with storage that we need to run a modern electricity grid, but we’re just not 3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 7/8 really quite there yet,” he said.

Rep. Salud Carbajal, a Democrat representing the area, has yet to take a position on the plant’s extension. But he urged Newsom to consider the local implications of keeping it open, including seismic safety.

Residents on the Central Coast had tussled for decades over the role of Diablo Canyon, but until recently, they thought the debate was over. Employees made retirement plans. Construction at the plant slowed as PG&E began preparing to shut the twin reactors down. Conservationists eyed the lands around the plant.

An extension would upset those plans.

In an interview, Carbajal said he’s also wondered whether it will get in the way of developing nearby offshore wind farms.
“I’m not one of those people that’s gonna say, ‘oh, you know, done and over,’” he said, referring to the debate over Diablo Canyon. “There’s a lot of moving parts.”
By 2045, state officials want offshore wind to produce 25 percent of California’s power, a more ambitious goal than any other state.

But even the most optimistic people acknowledge that those wind turbines won’t start producing power for the grid until at least 2029 — right before Diablo Canyon’s new closure date.

FILED UNDER: CLIMATE CHANGE, CALIFORNIA, GAVIN NEWSOM, HEAT WAVE, NUCLEAR POWER Huddle
A play-by-play preview of the day’s congressional news

SIGN UP
By signing up you agree to allow POLITICO to collect your user information and use it to better recommend content to you, send you email newsletters or updates from POLITICO, and share insights based on aggregated user information. You further agree to our privacy policy and terms of service. You can unsubscribe at any time and can contact us here. This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

EMAIL
Your Email Loading
About Us
Advertising
Breaking News Alerts
Careers
3/16/23, 11:31 AM California's latest power grid problems are just the beginning - POLITICO https://www.politico.com/news/2022/09/23/californias-lofty-climate-goals-clash-with-reality-00058466 8/8 Credit Card Payments
Digital Edition
FAQ
Feedback
Headlines
Photos
POWERJobs
Press
Print Subscriptions
Request A Correction
Write For Us
RSS
Site Map
Terms of Service
Privacy Policy
Do not sell my info
Notice to California Residents
© 2023 POLITICO LLC

i

David Shabazian, Director
Uduak-Joe Ntuk, California State Oil and Gas Supervisor California Department of Conservation
801 K Street, MS 24-01
Sacramento, CA 95814

October 1, 2021

RE: Response to CalGEM Questions for the California Oil and Gas Public Health Rulemaking Scientific Advisory Panel

Director Shabazian and Supervisor Ntuk,

Please find attached the responses from the California Oil and Gas Public Health Rulemaking Scientific Advisory Panel to the written questions sent by the California Geologic Energy Management Division (CalGEM) on August 31, 2021.

We would be glad to answer any further questions that may arise.

Best Regards,

Seth B.C. Shonkoff, PhD, MPH
Co-Chair, California Oil and Gas Public Health Rulemaking Scientific Advisory Panel Executive Director, PSE Healthy Energy
Visiting Scholar, Department of Environmental Science, Policy, and Management, University of California, Berkeley
Affiliate, Energy Technologies Area, Lawrence Berkeley National Lab

Rachel Morello-Frosch, PhD, MPH
Co-Chair, California Oil and Gas Public Health Rulemaking Scientific Advisory Panel Professor, Department of Environmental Science, Policy and Management & School of Public Health, University of California, Berkeley, Berkeley CA

Joan A. Casey, PhD, MA
Assistant Professor, Department of Environmental Health Sciences, Columbia University Mailman School of Public Health, New York, New York

Nicole Deziel, PhD, MHS
Associate Professor, Department of Environmental Health Sciences, Yale School of Public Health, Yale University, New Haven, Connecticut

ii
Dominic C. DiGiulio, PhD, MS
Senior Research Scientist, PSE Healthy Energy
Affiliate, Department of Civil, Environmental, and Architectural Engineering, University of Colorado, Boulder

Stephen Foster, PhD
Senior Principal, Geosyntec Consultants

Robert Harrison, MD and MPH
Clinical Professor of Medicine, Division of Occupational and Environmental Medicine, University of California San Francisco

Jill Johnston, PhD, MS
Assistant Professor of Environmental Health, Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California

Kenneth Kloc, PhD and MPH
Staff Toxicologist, Office of Environmental Health Hazard Assessment, California EPA

Lisa McKenzie, PhD and MPH
Clinical Assistant Professor, Department of Environmental and Occupational Health, Colorado School of Public Health, University of Colorado Denver Anschutz Medical Campus

Thomas McKone, PhD
Professor Emeritus, School of Public Health, University of California, Berkeley Affiliate, Energy Technologies Area, Lawrence Berkeley National Laboratory

Mark Miller, MD, MPH
Director, Children’s Environmental Health Center, Office of Environmental Health Hazard Assessment, California EPA
Associate Clinical Professor, Division of Occupational and Environmental Medicine, University of California, San Francisco

Andrea Polidori, PhD
Advanced Monitoring Technologies Manager, South Coast Air Quality Management District

1 CalGEM Questions for the California Oil and Gas Public Health Rulemaking Scientific Advisory Panel
CalGEM requests the California Oil and Gas Public Health Rulemaking Scientific Advisory Panel assistance with the following questions:

1. How would the panel characterize the level of certainty that proximity to oil and gas extraction wells and associated facilities in California causes negative health outcomes? Is there a demonstrated causal link between living near oil and gas wells and associated facilities and health outcomes?

We have focused our review on epidemiological studies carried out in multiple oil and gas regions, including Colorado, which has a similar regulatory context as California. Given that similar environmental health hazards and risks are intrinsic to both con ventional and unconventional oil and gas development (OGD), including exposure pathways, chemicals associated with hydrocarbon reservoirs, use of ancillary equipment, and non -chemical stressors (See section on “Similarities and Differences Between Unconven tional and Conventional OGD”), the California Oil and Gas Public Health Rulemaking Scientific Advisory Panel (Panel) concludes that the full body of epidemiologic literature is relevant to assess the human health hazards, risks and impacts of upstream OGD in California.

Our Panel concludes with a high level of certainty 1 that the epidemiologic evidence indicates that close residential proximity to OGD is associated with adverse perinatal and respiratory outcomes, for which the body of human health studies is most extensive in California and other locations.

Studies on Oil and Gas Development and Perinatal Outcomes Perinatal outcome studies provide the largest [19 studies] 2 and strongest body of evidence linking OGD exposure during the sensitive prenatal period with adverse health effects. The majority of studies that examine perinatal effects found increased risk of adverse birth outcomes in those most exposed to OGD (measured using metrics including, but not limited to proximity, well density, and production volume). It should also be noted that adverse perinatal outcomes, including preterm births, low birth weight, and small -for-gestational age births

1 In this document, the statement, “a high-level of certainty” is based on the professional judgement of all California Oil and Gas Public Health Rulemaking Scientific Advisory Panel (Panel) members in their assessment of the scientific evidence. In terms of panel process, all Panel members agree with the responses to the questions in this document. Any Panel member could have written a dissenting opinion, but no one requested to do so. This document reflects the perspective of the Panel members and not necessarily the opinions of their employers or institutions.

2 Apergis et al., 2019; Busby & Mangano, 2017; Caron-Beaudoin et al., 2020; Casey et al., 2016; Currie et al., 2017; Cushing et al., 2020; Gonzalez et al., 2020; Hill, 2018; Janitz et al., 2019; Ma, 2016; McKenzie et al., 2014, 2019; Stacy et al., 2015; Tang et al., 2021; Tran et al., 2020, Forthcoming; Walker Whitworth et al., 2018; Whitworth et al., 2017; Willis et al., 2021.

2 increase the risk of mortality and long -term developmental problems in newborns (Liu et al., 2012; Vogel et al., 2018) as well as longer term morbidity through adulthood (Baer et al., 2016; Barker, 1995; Carmody & Charlton, 2013; Frey & Klebanoff, 2016).

Perinatal Outcomes Associated with Conventiona l and Unconventional Oil and Gas Development
While many perinatal outcome studies outside of California focus on unconventional OGD (e.g., high-volume hydraulic fracturing), a recent review of the literature (Deziel et al., 2020) , highlighted the need for an updated assessment of the health effects associated with OGD more generally, as both conventional and unconventional OGD operations present health risks, especially to those living in close proximity. This bolsters conclusions reached by the authors of the 2015 independent scientific study of hydraulic fracturing and well stimulation in California led by the California Council on Science and Technology (CCST) (Long et al., 2015) pursuant to Senate Bill 4 (2013, Pavley). Recent studies in California have reported associations between exposure to OGD and adverse birth outcomes, considering wells under production using enhanced oil recovery including cyclic steam injection, steam flooding and water flooding -- methods that do not meet the definition of unconventional development (Gonzalez et al., 2020; Tran et al., 2020, Forthcoming). Similar findings regarding adverse birth outcomes have been reported while examining unconventional OGD in Colorado, Oklahoma, Pennsylvania and Texas (Apergis et al., 2019; Casey et al., 2016; Cushing et al., 2020; Gonzalez et al., 2020; Hill, 2018; McKenzie et al., 2019; Stacy et al., 2015; Walker Whitworth et al., 2018; Whitworth et al., 2017) . In the California independent scientific study on well stimulation pursuant to Senate Bill 4 (2013, Pavley), the authors concluded that while hydraulic fracturing introduces some specific human health risks, the majority of environmental risks and stressors are similar across conventional and unconventional oil and gas operations (Long et al., 2015; Shonkoff et al., 2015). Further, a handful of epidemiological studies explicitly examine potential differences in associations between conventional or unconventional oil or natural gas development and adverse outcomes. For example, Apergis et al. (2019) reported statistically significant reductions in infant health index within 1 k m of both conventional and unconventional drilling sites in Oklahoma. In summary, the Panel concludes with a high level of certainty that human health studies focused on unconventional and conventional OGD are relevant to consider in the California context where conventional development is most prevalent.

Consistency Across Perinatal Epidemiology Studies
We have a high level of certainty in the findings in the body of epidemiological studies for perinatal health outcomes because of the consistency of results across multiple studies that were conducted using different methodologies, in different locations, with diverse populations, and during different time periods (see Table 1 below). Most of these studies entail rigorous, high quality analyses (i.e., study designs that establish temporality based on large sample sizes, control for potential individual and area -level confounders, apply rigorous statistical

3 modelling techniques, and conduct sensitivity analyses to assess the robustness of effects). A variety of pollutants (e.g., PM 2.5 and air toxics) and other OGD stressors are associated with these same adverse birth outcomes (Dzhambov & Lercher, 2019; Nieuwenhuijsen et al., 2017; Shapiro et al., 2013) , which further strengthens the evidence of the link between OGD and adverse perinatal outcomes. Therefore, the totality of the epidemiological evidence provides a high level of certainty that exposure to OGD (and associated exposures) cause a significant increased risk of poor birth outcomes.

Further, imprecision in exposure assessment or non -differential exposure misclassification in some of the epidemiological studies is more likely to attenuate observed relationships, thus leading to an underestimate of the true adverse impacts of OGD on birth outcomes (Figure 1).

In environmental epidemiologic studies, researchers often use surrogates to estimate exposures or assign individuals to exposure categories; these surrogates have some measurement error associated with them. When these errors in ass igning or classifying participant exposures are similar between exposed and unexposed or those with or without the health outcome, this is referred to as non -differential exposure misclassification. This type of “noise” in the data tends to dilute or atten uate the true exposure -response relationship, as illustrated by the hypothetical dashed line in Figure 1, which has a shallower slope compared to the hypothetical “true” solid line.

Figure 1. Effect of imprecise exposure estimates on a hypothetical expos ure-response relationship (Source: Adapted from Seixas & Checkoway, 1995).

4 Respiratory Risks and Impacts from Oil and Gas Development Respiratory health outcomes are the second most studied health outcomes in the epidemiological literature examining OGD, with eight peer-reviewed studies published to date.

Two peer-reviewed studies in California found an association between OGD and self-reported and physician-diagnosed asthma, reduced lung function, and self -reported acute respiratory symptoms (e.g., recent wheeze) (Johnston et al., 2021; Shamasunder et al., 2018). Six studies in other oil and gas regions (Pennsylvania and Texas) reported an association between OGD and asthma exacerbations, asthma hospitalizations, and respiratory symptoms (Koehler et al., 2018; Peng et al., 2018; Rabinowitz et al., 2015; Rasmussen et al., 2016; Willis et al., 20 18, 2020).

Epidemiological studies, by design, often use aggregate measures of exposure to account for multiple potential stressors and pathw ays associated with OGD (e.g., air pollution, noise pollution, groundwater and/or drinking water contamination). Many criteria air pollutants (e.g., particulate matter, ozone, nitrogen oxides) and hazardous air pollutants emitted from OGD have a well-established body of scientific literature indicating that exposure to these pollutants causes an increased risk of development and exacerbation of respiratory disease (Bolden et al., 2015; Ferrero et al., 2014). We reiterate the relevance of studies on both conventional and unconventional OGD for respiratory health outcomes. For example, (Willis et al., 2020) found that both conventional and unconventional natural gas development at the ZIP code level was associated with pediatric asthma hospitalizations in Texas.

Comparing The Body of Perinatal and Respiratory Outcome Studies Against The Bradford Hill Criteria for Causation
Below, we demonstrate how the body of epidemiological studies on the relationship between OGD and perinatal and respiratory outcomes meets the nine Bradford Hill Criteria for Causation (Hill, 1965; Lucas & McMichael, 2005) . The Bradford Hill Criteria are used to evaluate the strength of epidemiological evidence for determining a causal relationship between an exposure and observed effect. These criteria are widely used in the field of epidemiology and public health practice to guide decision -making. After considering these criteria, the Panel concludes with a high level of certainty that there is a causal relationship between close geographic proximity to OGD and adverse perinatal and respiratory outcomes (Table 1).

5 Table 1. Application of the Bradford Hill Criteria for Causation to the peer-reviewed epidemiological literature on oil and gas development and perinatal and respiratory health outcomes.

Criteria for Causation
(Bradford-Hill)
Description of
Criteria Perinatal Health Studies Respiratory Health Studies Strength of
Association
Environmental studies
commonly report
modest effects sizes
(i.e., relative to active
tobacco smoking or
alcohol consumption).

A small magnitude of
association can
support a causal
relationship, a larger
association may be
more convincing.

Reported effect sizes are in ranges
similar to other well-established
environmental reproductive and
developmental hazards, such as PM2.5
(Dadvand et al., 2013; C. Li et al.,
2020). Some studies, particularly those
in California, have found stronger
effect estimates for OGD exposures
among socially marginalized groups
(Cushing et al., 2020; Gonzalez et al.,
2020; Tran et al., 2020, Forthcoming).

Reported effect sizes are in ranges similar
to other well-established environmental
respiratory hazards. For example, effect
sizes in reductions in lung function by
Johnston et al. (2021) are similar in
magnitude to reductions in lung function
associated with secondhand smoke
exposure among women (Eisner, 2002)
and reductions in lung function among
adults living near busy roadways (e.g.,
(Kan et al., 2007).

Consistency Consistent findings
observed by different
persons in different
places with different
samples strengthens
the likelihood of an
effect.

Adverse birth outcomes have been
observed in multiple studies using
multiple methods in different
populations at different times and
locations (e.g., California,
Pennsylvania, Colorado, Texas). While
there is some variation in findings by
specific perinatal outcomes, the overall
body of evidence is highly consistent in
supporting the association between
OGD and adverse perinatal outcomes.

Various respiratory health outcomes are
evaluated in the literature. For asthma --
the most commonly studied respiratory
health outcome -- studies across
California, Pennsylvania and Texas
consistently show an association between
OGD and asthma-related metrics (asthma
prevalence, exacerbations, pediatric
hospitalizations) (Koehler et al., 2018;
Rasmussen et al., 2016; Shamasunder et
al., 2018; Willis et al., 2018, 2020) .

6 Criteria for Causation
(Bradford-Hill)
Description of
Criteria Perinatal Health Studies Respiratory Health Studies Specificity Causation is likely if
there is no other likely
explanation.

All peer-reviewed birth outcome
studies included in our review
controlled for other potential
confounders by (i) accounting or
adjusting for other individual-level or
area-level factors (e.g., other air
pollution sources, neighborhood
socioeconomic status) in the analysis
(Casey et al., 2016; McKenzie et al.,
2014; Tran et al., 2020, Forthcoming).

Other studies applied statistical
modeling approaches such as
difference-in-difference that accounts
for temporal and spatial trends that
may confound observed effects (Willis
et al., 2021).

Most respiratory health studies have
controlled for other potential explanatory or
confounding factors by (i) accounting or
adjusting for other individual-level (e.g.,
smoking status) or area-level factors (e.g.,
other air pollution sources) in the analysis
(Johnston et al., 2021; Koehler et al., 2018;
Peng et al., 2018; Rabinowitz et al., 2015;
Rasmussen et al., 2016; Willis et al., 2018,
2020), or in the study design, such as
utilizing a difference-in-difference
methodology (Peng et al., 2018; Willis et
al., 2018).

Temporality Exposure precedes the
disease.

Most birth outcomes studies have
proper temporal alignment between
exposure and outcome and use a
retrospective cohort, case control or
other study design that allows
retroactive assessment of exposures to
OGD occurring before the onset of
disease. They do not consider
exposure that occurred at the time of
disease or oil and gas wells drilled
after the disease.

Some respiratory health studies do not
allow for assessments of exposure that
predate disease. However, of the studies
with the proper temporal alignment
(Johnston et al., 2021; Koehler et al., 2018;
Peng et al., 2018; Rasmussen et al., 2016;
Willis et al., 2018), authors report
statistically significant associations
between OGD and oral corticosteroid
medication orders, asthma hospitalizations
and asthma-related emergency department
visits.

7 Criteria for Causation
(Bradford-Hill)
Description of
Criteria Perinatal Health Studies Respiratory Health Studies Biological Gradient
(Dose-Response)
Greater exposure leads
to a greater likelihood
of the outcome.

Some studies have found dose-
response relationships based on oil
and gas production volume categories
or metrics of inverse distance
weighting and/or oil and gas well
density in California and elsewhere
(Casey et al., 2016; McKenzie et al.,
2014, 2019; Tang et al., 2021; Tran et
al., 2020).

Larger reductions in lung function observed
with decreased distance from active oil
development sites (Johnston et al., 2021).

Plausibility The exposure pathway
and biological
mechanism is plausible
based on other
knowledge.

Individual health-damaging chemical
pollutants are well-understood to be
emitted from OGD (e.g., PM2.5,
benzene) and established as
contributing to increased risk for the
same adverse perinatal outcomes
observed in the epidemiology studies.

Stressors associated with OGD (e.g.,
psychosocial stress; (Casey et al.,
2019) can also contribute to increased
adverse perinatal outcomes.

Many air pollutants associated with OGD
are well-known to contribute to respiratory
morbidity and mortality, including
exacerbations of existing respiratory
conditions (Guarnieri & Balmes, 2014).

Coherence Causal inference is
possible only if the
literature or substantive
knowledge supports
this conclusion.

In particular, the body of peer-reviewed
literature is converging towards
singular directions for adverse
perinatal outcomes.

The body of peer-reviewed literature points
in a singular direction for adverse
respiratory health outcomes.

8 Criteria for Causation
(Bradford-Hill)
Description of
Criteria Perinatal Health Studies Respiratory Health Studies Experiment Causation is a valid
conclusion if
researchers have seen
observed associations
in prior experimental
studies.

N/A- Human population-based
experimental studies are not available
due to ethical issues.

N/A- Human population-based
experimental studies are not available due
to ethical issues.

Analogy For similar programs
operating, similar
results can be
expected to bolster the
causal inference
concluded.

Pollutants well known to be emitted
during OGD including benzene,
toluene and 1,3 butadiene are listed as
reproductive or developmental
toxicants under Prop 65 and thus are
recognized as such by the State of
California (CalEPA OEHHA, 2021).

EPA’s current Integrated Science
Assessments of particulate matter and
tropospheric ozone conclude that the
evidence is suggestive of, but is not
sufficient to infer, a causative
relationship between birth outcomes,
including preterm birth and low birth
weight, and PM2.5 and long term ozone
exposures (US EPA, 2019, 2020).

Additionally, increased stress during
pregnancy can alter fetal growth and
length of gestation (Fink et al., 2012).

EPA’s current Integrated Science
Assessments of particulate matter and
tropospheric ozone conclude that there is:
a casual relationship between respiratory
outcomes, including asthma and short term
ozone exposure; and likely a causal
relationship between respiratory outcomes,
including asthma and: short and long term
PM2.5 exposure; and long term ozone
exposure (US EPA, 2019, 2020).

9 Similarities and Differences Between Unconventional and Conventional Oil and Gas Development

Though definitions of conventional and unconventional OGD may differ across different regulatory and policy landscapes, the majority of OGD in California is often considered conventional, involving vertical drilling at shallower depths into target geologies that hold migrated hydrocarbons. These attributes of development are often considered in contrast to unconventional OGD, which can involve horizontal directional drilling in deeper wells to access source rock formati ons by increasing the permeability of these tight formations using mostly hydraulic fracturing. In addition, these unconventional operations are often accompanied with greater masses of material inputs (e.g., water, chemical additives, proppants) and a gre ater magnitude of liquid and solid waste outputs (e.g., flowback fluids and produced water). It should be noted, however, that hydraulic fracturing that takes place in California often uses fluids (gels) with higher concentrations of well stimulation chemicals than those fluids used in high-volume slick water hydraulic fracturing of source rock in other parts of the United States (Long et al., 2015).

However, many environmental and health hazards and risks are intrinsic to both conventional and unconventional OGD (Hill et al., 2019; Jackson et al., 2014; Lauer et al., 2018; Stringfellow et al., 2017; Zammerilli et al., 2014) . PM2.5 and nitrogen oxides emissions result from the use of diesel -powered equipment and trucks and hazardous air pollutants such as benzene, toluene, ethylbenzene and xylene (BTEX) occur naturally in oil and gas formations, regardless of the type of extrac tion method employed. Noise pollution, odors, and landscape disruption are inherent to OGD. Investigations in other oil and gas states have noted radioactivity on particles downwind from unconventional oil and gas wells (Li et al., 2020b) and in sediment downstream of water treatment plants that treat waste from conventional as well as unconventional oil and gas operations (Burgos et al., 2017; Lauer et al., 2018).

In California, policy, regulatory and scientific emphasis has been placed on well stimulation activities, including hydraulic fracturing, matrix acidizing and acid fracturing. The 2015 Independent Scientific Assessment on Well Stimulation in California, wh ich focused primarily on well stimulation activities pursuant to Senate Bill 4 (2013, Pavley), reported the following key conclusion: “The majority of impacts associated with hydraulic fracturing are caused by the indirect impacts of oil and gas production enabled by the hydraulic fracturing” (Long et al., 2015). Indirect impacts relevant to human health for the purpo ses of the study included:
“proximity to any oil production, including stimulation - enabled production, could result in hazardous emissions to air and water, and noise and light pollution that could affect public health” (Long et al., 2015). Additionally, a recent evaluation of chemical usage during OGD in California found significant overlap in chemical additives use d for well stimulation (including hydraulic fracturing) and those used in routine activities, such as well maintenance (Stringfellow et al., 2017).

10 2. What are the air pollutants released from these activities that cause negative health outcomes? How do we know exposure to these is likely from oil and gas extraction wells and associated facilities, as opposed to other sources?

The wells, valves, tanks and other equipment used to produce, store, process and transport petroleum products at both unconventional and conventional OGD sites are associated with emissions of toxic air contaminants, hazardous air pollutants and other health -damaging non- methane VOCs (Helmig, 2020; Moore et al., 2014) . Diesel engines used to power on -site equipment and trucks at unconventional and conventional OGD sites directly emit health - damaging hazardous air pollutants, fine particulate matter (PM2.5), nitrogen oxides and volatile organic compounds (VOCs) (CalEPA OEHHA, 2001) . Many VOCs and nitrogen oxides are precursors to ground level ozone (O 3) formation, another known health harming pollutant.

Hazardous air pollutants that are known to be emitte d from OGD sites include benzene, toluene, ethylbenzene, xylenes, hexane and formaldehyde --many of which are known, probable or possible carcinogens and/or teratogens and which have other adverse effects for non-cancer health outcomes (CalEPA OEHHA, 2008, 2009; Moore et al., 2014) . In the San Joaquin Valley Air Pollution Control District, OGD activities are responsible for the majority of emissions of multiple toxic air contaminants including acetaldehyde, benzene, formaldehyde, hexane and hydrogen sulfide (Figure 2) (Brandt et al., 2015; Long et al., 2015).

Figure 2. Toxic Air Contaminant emissions from stationary facilities in the San Joaquin Valley Air Pollution Control District (Source: (Brandt et al., 2015).

11 A recently published study using statewide air quality monitoring data from California investigated whether drilling new wells or increasing production volume at active wells resulted in emissions of PM2.5, nitrogen dioxide (NO2), VOCs, or O3 (Gonzalez et al., 2021). To assess the effect of oil and gas activities on concentrations of air pollutants, the authors used daily variation in wind direction as an instrumental variable and used fixed effects regression to control temporal factors and time-invariant geographic factors. The authors documented higher concentrations of PM2.5, NO2, VOCs, and O3 at air quality monitoring sites within 4 km of pre - production OGD well sites (i.e., wells that were between spudding and completion) and 2 km of production OGD well sites, after adjusting for geographic, meteorological, seasonal, and time trending factors. In placebo tests, the authors assessed exposure to well sites downwind of the air monitors and observed no effect on air pollutant concentrations. Table 2 summarizes the increases in each pollutant for each additional upwind well site by distance.

Table 2. Summary of air pollutant concentrations measured between 2006 -2019 at 314 air quality monitoring sites in the EPA Air Quality System for California (Gonzalez et al., 2021).

Distance PM2.5 µg/m3* NO2 ppb VOCs (ppb C)* O3 (ppb)
Estimated increase for each additional upwind pre-production well site Within 2 km 2.35 (0.81, 3.89) 2.91 (0.99, 4.84) No increase no increase 2-3 km 0.97 (0.52, 1.41) 0.65 (0.31, 0.99) No increase 0.31 (0.2, 42) 3-4 km no increase no increase no increase 0.14 (0.05, 0.23) Estimated Increase for each 100 BOE of total oil and gas upwind production volume 1 km 1.93 (1.08, 2.78) 0.62 (0.37, 0.86) 0.04 (0.01, 07) no increase 1-2 km no increase no increase no increase 0.11 (0.08, 0.14) *No PM2.5 or VOC monitoring sites with 1 km of pre -production well sites; BOE, barrels of oil equivalents.

These multiple stressors, along with other physical factors such as noise and vibration, are consistently found in exposure studies to be measurabl y higher near oil and gas extraction wells and other ancillary infrastructure in California. As such, the Panel concludes with a high level of certainty that concentrations of health -damaging air pollutants, including criteria air pollutants and toxic air contaminants, are more concentrated near OGD activities compared to further away.

12 3. Does the evidence evaluated clearly support a specific setback? If so, what is this setback distance and what oil and gas extraction activities would it specifically apply to? What is the supporting evidence?
a. How does this evidence justify the recommended setback distance, as opposed to another distance?
Existing epidemiologic studies were not designed to test and establish a specific “safe” buffer distance between OGD sites and sensitive receptors, such as homes and schools.

Nevertheless, studies consistently demonstrate evidence of harm at distances less than 1 km, and some studies also show evidence of harm linked to OGD activity at distances greater than 1 km. In addition, exposure pathway studies have demonstrated through measurements and modelling techniques, the potential for human exposure to numerous environmental stressors (e.g., air pollutants, water contaminants, noise) at distances less than 1 km (e.g., Allshouse et al., 2019; Holder et al., 2019; McKenzie et al., 2018; DiGiulio et al., 2021; Soriano et al., 2020), and that the likelihood and magnitude of exposure decreases with increasing distance.

b. What are the health benefits from this setback? Can the panel quantify them or recommend a methodology CalGEM can use to quantify them? Can the panel establish that these health benefits can only be achie ved with the setback? Or can they also be achieved with mitigation controls?

Figure 3 presents a hierarchy of strategies to reduce human health hazards, risks and impacts from OGD activities. Table 3 presents the advantages and disadvantages of each strategy from an environmental public health perspective.

Figure 3. Hierarchy of strategi es to reduce or eliminate public health harms for OGD activities. Note: the use of the term “wells” includes the ancillary infrastructure used to develop, gather and process oil and gas in the upstream oil and gas sector.

13 At the top of Figure 3 is the most health protective strategy: to stop drilling and developing new wells, phase out existing OGD activities and associated infrastructure, and properly plug remediate legacy wells and ancillary infrastructure.

If the development of oil and gas is to contin ue, the greatest health benefits would be gained from a strategy that includes the next two controls in the hierarchy depicted in Figure 3: the elimination of new and existing wells and ancillary infrastructure within scientifically informed setback distan ces and the deployment of engineering emission controls and associated monitoring approaches that lead to rapid leak detection and repair for new and existing wells and ancillary infrastructure. Because air pollutant concentrations and noise levels decreas e with increasing distance from a source, adequate setbacks can reduce harm to local populations by reducing exposures to air pollutants and noise directly emitted from the OGD activities. However, setbacks do not reduce harms from OGD contributions to reg ional air pollutant levels, such as secondary particulate matter and ozone, or greenhouse gases, such as methane, which are nearly always co -mingled with health -damaging air pollutants (Michanowicz et al., Forthcoming). Engineering controls that reduce emissions at the well site are also necessary to reduce these harms.

Engineering controls include cradle -to-grave noise and air pollution emission mitigation controls on OGD infrastructure including new, modified and existing infrastructure, and proper abandonment of legacy infrastructure, prioritizing those nearest to residential sites and schools and those associated with the highest emissions, leaks and other environmental hazards.

However, engineering controls can fail and engineering solutions may not be available for or economically feasible to handle all of the complex stressors generated by OGD, including multiple sources and types of air pollution, noise pollution, light pollution, water pollution, and other stress ors. Therefore, neither setbacks or engineering controls alone are sufficient to reduce the health hazards and risks from OGD activities -- both approaches are needed in tandem.

Finally, we note that while outside of CalGEM’s jurisdiction, setbacks for new construction of housing or schools at a certain distance from existing or permitted OGD sites (commonly referred to as reverse setbacks), should be considered.

14 Table 3. Advantages and Disadvantages of Oil and Gas Development Control Strategies from an Environmental Public Health Perspective.

Control Strategy Description Advantage Disadvantage
Elimination Eliminate or reduce
new and existing wells
and ancillary
infrastructure in
combination with
proper plugging and
abandonment of wells
and other legacy
infrastructure.

Eliminates the source of
nearly all environmental
stressors (e.g., air and
water pollutants, noise);
protects local and regional
populations
None.

Setbacks Increase the distance
between OGD
hazards and sensitive
receptors.

Reduces risk of exposures
to populations living near
OGD sites; environmental
stressors are generally
attenuated with increasing
distance.

Setbacks alone without coupled
engineered mitigation controls
allow continued release of
hazards and therefore does not
adequately address air pollutant
and greenhouse gas emissions
from OGD and their impacts on
regional air quality and the
climate.

Engineering
Controls
Reduces or eliminates
release of specific
hazards on site.

Reduces or eliminates
certain hazards and
therefore can have local
and regional
environmental public
health benefits.

Tends to be disproportionately
focused on air pollutant
emissions. Often not feasible to
apply engineering solutions to
multiple, complex stressors
each requiring different control
technologies (e.g. noise, air and
water impacts, social stressors)
and lacks the important factor of
safety provided by a setback
when engineering controls fail.

Residence
Controls
Provides households
with devices to reduce
hazard at the home
(e.g., water filter, light-
blocking shades, air
filters).

Reduces intensity of
certain hazards to nearby
communities at the
household level.

Places burden on individuals
and households to use devices
properly and to maintain and
regularly replace controls to
maximize effectiveness. Not
feasible to apply devices to
address numerous, complex
stressors.

Personal
Protective
Equipment
Provide individuals
with devices to reduce
exposure (e.g.,
respiratory masks, ear
plugs, eye masks).

Reduces intensity of
exposure of certain
hazards to nearby
individuals.

Places burden on individuals to
use PPE consistently and
properly and is not feasible for
the complex stressors.

15 Attributable Risk Calculations

One method to estimate health harms from OGD is to use the measures of association from the epidemiologic literature and population counts to calculate the excess number of specific health outcomes. This is what is known as an attributable risk method. We may be able to derive these estimates in the final report for birth outcomes using estimates of population counts for women of reproductive age in California living near OGD sites. We will also attempt to derive similar estimates for respiratory outcomes by using age appropriate population counts near OGD sites. This attributable risk method can allow us to estimate the number of adverse perinatal or respiratory cases that are attributable to OGD exposures and could be attenuated through the implementation of elimination or setback strategies.

c. Can the panel quantify or recommend a methodology CalGEM can use to quantify the health benefits associated with mitigation controls?

The Panel was not tasked to estimate health benefits of various setbacks and mitigation strategies, which pose significant methodological challenges and would require considerab le time and effort. Among the challenges is the need to consider the benefits of reducing multiple stressors -- multiple air pollutants and other chemicals, noise, vibration, light, subsurface contamination, etc.

Known Health Benefits of Reducing Air and Noise Pollution

There is a significant body of literature and available tools that address the potential health benefits that can be achieved by reducing air and noise pollution exposures. The National Institute of Environmental Health Sciences has linke d air pollution and specifically PM 2.5 to respiratory disease, cardiovascular disease, cancer, and reproduction harm and provides references supporting these links (NIEHS (National Institute of Environmental Health Sciences), 2021) . Schraufnagel et al. (2019) examined in detail the health benefits of air pollution reductions in different geographic regions. Friedman et al. (2001) showed that improvements in air quality in preparation for the 1996 Atlanta Olympics resulted in significantly lower rates of childhood asth ma events, including reduced emergency department visits and hospitalizations. Avol et al. (2001) demonstrated that children in southern California who moved to communities with higher air pollution levels had lower lung function growth rates than children who moved to areas with lower air pollution levels.

Gauderman et al. (2015), examining the impact of reductions in PM2.5 and nitrogen dioxide in the Los Angeles air basin, found that children who grew up after air quality improvements had less than ½ the chance of having clinically low lung function results. Ha et al. (2014) found PM2.5 exposures in all trimesters to be significantly and positively associated with the risk of all adverse birth outcomes.

16 In an analysis of noise exposure reductions. Based on sound levels measured and/or modeled across the US together with an EPA exposure - response model for levels exceeding EPA standards, Swinburn et al. (2015) found that a 5-dB noise reduction scenario in communities with noise exceeding EPA standards would reduce the prevalence of hypertension by 1.4% and coronary heart disease by 1.8%. The types of health-benefit studies noted here provide a basis for conducting a health-benefits analysis using a tool such as US EPA’s Environmental Benefits Mapping and Analysis Program—Community Edition (BenMAP-CE) (US EPA, 2021).

Possible Approaches to Quantify Health Benefits

CalGEM could obtain estimate s of the health benefits achieved from different mitigation strategies individually or in combination with tools such as the Community Multiscale Air Quality Model (CMAQ) (Binkowski & Roselle, 2003) and/or other exposure assessment tools and link model output to EPA’ s BenMAP-CE (US EPA, 2021) . However, these models and approaches are only focused on air quality and noise. It should also be noted that a significant drawback of using BenMAP -CE for this application is that it only considers impacts from criteria air pollutants and not from toxic air contaminants or other emerging air pollutants.

BenMAP-CE estimates the number and economic value of health impacts resulting from changes in air pollution concentrations. BenMAP -CE estimates benefits in terms of the reductions in the risk of premature death, heart attacks, and other adverse health effects.

BenMAP-CE requires as input, pollutant concentrations at a scale that matches with population data. These concentrations can be obtained from a model such as CMAQ (Binkowski & Roselle, 2003) or from a monitoring network. BenMAP -CE takes the concentration fields for a base case and then for a pollution reduction (or increase) to assess health benefits (or detriments). BenMAP -CE then estimates changes in health endpoints, allowing the user to specify the concentration –response function and either use built -in population and baseline mortality rates or specify them as inputs.

It should be noted that in order to use a model such as BenMAP-CE to assess health benefits of setbacks and mitigation controls at well sites across California would involve a significant level of time and effort in data collection and model executions. In addition, these models are limited to characterizing the heal th benefits of criteria air pollutant reductions, but do not account for other OGD related exposures such as toxic air contaminants, other chemical exposures and exposures to other stressors through other environmental pathways (e.g., water and noise). Add itionally, and importantly, the lack of spatially resolved emissions data from upstream OGD introduces challenges when assessing local - and sub-regional scaled health impacts that would be required for calculating benefits of specific policies such as setbacks and emission control. As such, attempts to quantify benefits using BenMAP-CE are likely to underestimate them.

17 4. CalGEM is aware of health risk assessments, health impact assessments, air exposure studies, and workforce safety studies that have been c onducted but were not evaluated as part of your preliminary advice. How do these studies align with your causation determination, any recommended setback distance, and recommendations on health benefits quantification?
The Panel determined early in its de liberations that it would limit the studies assessed in its report to those in the peer-reviewed scientific literature. This criterion ensures that studies have been evaluated by scientists who have not been involved with the study but have expertise in the relevant topic area and/or the methods used to carry out analyses, prior to publication. The peer-review process helps to ensure that high quality data and scientific interpretations are at the core of the science -policy decision-making process. Authors of peer reviewed studies are more likely to have been questioned about their methods, data interpretations, and conclusions, leading to greater confidence in the results.

In addition, the Panel was not tasked with assessing occupational studies. If CalGEM staff are aware of any peer -reviewed studies that were not included in our preliminary advice, we encourage them to send the Panel references so that we can evaluate them for inclusion in the final report. We intend to scan the literature again to assess whether relevant studies have been published since we completed the draft report. Should additional peer -reviewed studies be identified, the Panel will evaluate them to determine if they align with the scope of the report and should be added.

18 References
Allshouse, W. B., McKenzie, L. M., Barton, K., Brindley, S., & Adgate, J. L. (2019).

Community Noise and Air Pollution Exposure During the Development of a Multi-Well Oil and Gas Pad. Environmental Science & Technology, 53(12), 7126–7135. https://doi.org/10.1021/acs.est.9b00052
Apergis, N., Hayat, T., & Saeed, T. (2019). Fracking and infant mortality: Fresh evidence from Oklahoma. Environmental Science and Pollution Research, 26(31), 32360–32367. https://doi.org/10.1007/s11356-019-06478-z
Avol, E. L., Gauderman, W. J., Tan, S. M., London, S. J., & Peters, J. M. (2001). Respiratory effects of relocating to areas of differing air pollution levels. American Journal of Respiratory and Critical Care Medicine, 164(11), 2067–2072. https://doi.org/10.1164/ajrccm.164.11.2102005
Baer, R. J., Rogers, E. E., Partridge, J. C., Anderson, J. G., Morris, M., Kuppermann, M., Franck, L. S., Rand, L., & Jelliffe-Pawlowski, L. L. (2016). Population-based risks of mortality and preterm morbidity by gestational age and birth weight. Journal of Perinatology, 36(11), 1008–1013. https://doi.org/10.1038/jp.2016.118 Barker, D. J. P. (1995). Fetal origins of coronary heart disease. BMJ, 311(6998), 171–174. https://doi.org/10.1136/bmj.311.6998.171
Binkowski, F. S., & Roselle, S. J. (2003). Models-3 Community Multiscale Air Quality (CMAQ) model aerosol component 1. Model description. Journal of Geophysical Research:

Atmospheres, 108(D6). https://doi.org/10.1029/2001JD001409 Bolden, A. L., Kwiatkowski, C. F., & Colborn, T. (2015). New Look at BTEX: Are Ambient Levels a Problem? Environmental Science & Technology, 49(9), 5261–5276. https://doi.org/10.1021/es505316f
Brandt, A., Millstein, D., Jin, L., & Englander, J. (2015). Air Quality Impacts from Well Stimulation. In An Independent Scientific Assessment of Well Stimulation in California. https://ccst.us/wp-content/uploads/160708-sb4-vol-II-3-1.pdf Burgos, W. D., Castillo-Meza, L., Tasker, T. L., Geeza, T. J., Drohan, P. J., Liu, X., Landis, J.

D., Blotevogel, J., McLaughlin, M., Borch, T., & Warner, N. R. (2017). Watershed-Scale Impacts from Surface Water Disposal of Oil and Gas Wastewater in Western Pennsylvania. Environmental Science & Technology.
https://doi.org/10.1021/acs.est.7b01696
Busby, C., & Mangano, J. J. (2017). There’s a World Going on Underground—Infant Mortality and Fracking in Pennsylvania. Journal of Environmental Protection, 08(04), 381. https://doi.org/10.4236/jep.2017.84028
CalEPA OEHHA. (2001, May 21). Health Effects of Diesel Exhaust. OEHHA. https://oehha.ca.gov/air/health-effects-diesel-exhaust
CalEPA OEHHA. (2008). Technical Support Document for the Derivation of Noncancer Reference Exposure Levels Appendix D 2014 Update.
https://oehha.ca.gov/media/downloads/crnr/noncancertsdfinal.pdf

19 CalEPA OEHHA. (2009). Technical Support Document for Cancer Potency Factors: 88.

CalEPA OEHHA. (2021, March 19). The Proposition 65 List. OEHHA. https://oehha.ca.gov/proposition-65/proposition-65-list Carmody, J. B., & Charlton, J. R. (2013). Short-Term Gestation, Long-Term Risk: Prematurity and Chronic Kidney Disease. Pediatrics, 131(6), 1168–1179. https://doi.org/10.1542/peds.2013-0009
Caron-Beaudoin, É., Whitworth, K. W., Bosson-Rieutort, D., Wendling, G., Liu, S., & Verner, M.-A. (2020). Density and proximity to hydraulic fracturing wells and birth outcomes in Northeastern British Columbia, Canada. Journal of Exposure Science & Environmental Epidemiology. https://doi.org/10.1038/s41370-020-0245-z Casey, J. A., Goin, D. E., Rudolph, K. E., Schwartz, B. S., Mercer, D., Elser, H., Eisen, E. A., & Morello-Frosch, R. (2019). Unconventional natural gas development and adverse birth outcomes in Pennsylvania: The potential mediating role of antenatal anxiety and depression. Environmental Research, 177, 108598.
https://doi.org/10.1016/j.envres.2019.108598
Casey, J. A., Savitz, D. A., Rasmussen, S. G., Ogburn, E. L., Pollak, J., Mercer, D. G., & Schwartz, B. S. (2016). Unconventional natural gas development and birth outcomes in Pennsylvania, USA. Epidemiology (Cambridge, Mass.), 27(2), 163–172. https://doi.org/10.1097/EDE.0000000000000387
Currie, J., Greenstone, M., & Meckel, K. (2017). Hydraulic fracturing and infant health: New evidence from Pennsylvania. Science Advances, 3(12), e1603021. https://doi.org/10.1126/sciadv.1603021
Cushing, L. J., Vavra-Musser, K., Chau, K., Franklin, M., & Johnston, J. E. (2020). Flaring from Unconventional Oil and Gas Development and Birth Outcomes in the Eagle Ford Shale in South Texas. Environmental Health Perspectives, 128(7), 077003. https://doi.org/10.1289/EHP6394
Dadvand, P., Parker, J., Bell, M. L., Bonzini, M., Brauer, M., Darrow, L. A., Gehring, U., Glinianaia, S. V., Gouveia, N., Ha, E., Leem, J. H., van, den H. E. H., Jalaludin, B., Jesdale, B. M., Lepeule, J., Morello, -Frosch Rachel, Morgan, G. G., Pesatori, A. C., Pierik, F. H., … Woodruff, T. J. (2013). Maternal Exposure to Particulate Air Pollution and Term Birth Weight: A Multi-Country Evaluation of Effect and Heterogeneity. Environmental Health Perspectives, 121(3), 267–373. https://doi.org/10.1289/ehp.1205575 Deziel, N. C., Brokovich, E., Grotto, I., Clark, C. J., Barnett-Itzhaki, Z., Broday, D., & Agay- Shay, K. (2020). Unconventional oil and gas development and health outcomes: A scoping review of the epidemiological research. Environmental Research, 182, 109124. https://doi.org/10.1016/j.envres.2020.109124
DiGiulio, D., Rossi, R.J., Jaeger, J., Shonkoff, S.B.C., & Ryan, J.N. (2021). Vulnerability of Groundwater Resources Underlying Unlined Produced Water Ponds in the Tulare Basin of the San Joaquin Valley, California. In Review.

20 Dzhambov, A. M., & Lercher, P. (2019). Road Traffic Noise Exposure and Birth Outcomes:

An Updated Systematic Review and Meta-Analysis. International Journal of Environmental Research and Public Health, 16(14), 2522. https://doi.org/10.3390/ijerph16142522
Eisner, M. D. (2002). Environmental tobacco smoke exposure and pulmonary function among adults in NHANES III: Impact on the general population and adults with current asthma.

Environmental Health Perspectives, 110(8), 765–770.
https://doi.org/10.1289/ehp.02110765
Ferrero, A., Iñiguez, C., Esplugues, A., Estarlich, M., & Ballester, F. (2014). Benzene Exposure and Respiratory Health in Children: A Systematic Review of Epidemiologic Evidences. Journal of Pollution Effects & Control, 02(02). https://doi.org/10.4172/2375- 4397.1000114
Fink, N. S., Urech, C., Cavelti, M., & Alder, J. (2012). Relaxation during pregnancy: What are the benefits for mother, fetus, and the newborn? A systematic review of the literature. The Journal of Perinatal & Neonatal Nursing, 26(4), 296–306. https://doi.org/10.1097/JPN.0b013e31823f565b
Frey, H. A., & Klebanoff, M. A. (2016). The epidemiology, etiology, and costs of preterm birth.

Seminars in Fetal and Neonatal Medicine, 21(2), 68–73.
https://doi.org/10.1016/j.siny.2015.12.011
Friedman, M. S., Powell, K. E., Hutwagner, L., Graham, L. M., & Teague, W. G. (2001).

Impact of Changes in Transportation and Commuting Behaviors During the 1996 Summer Olympic Games in Atlanta on Air Quality and Childhood Asthma. JAMA, 285(7), 897–905. https://doi.org/10.1001/jama.285.7.897
Gauderman, W. J., Urman, R., Avol, E., Berhane, K., McConnell, R., Rappaport, E., Chang, R., Lurmann, F., & Gilliland, F. (2015). Association of improved air quality with lung development in children. The New England Journal of Medicine, 372(10), 905–913. https://doi.org/10.1056/NEJMoa1414123
Gonzalez, D. J. X., Francis, C. K., Shaw, G. M., Cullen, M. R., Baiocchi, M., & Burke, M. (2021). Upstream oil and gas production and ambient air pollution in California. Science of The Total Environment, 150298. https://doi.org/10.1016/j.scitotenv.2021.150298 Gonzalez, D. J. X., Sherris, A. R., Yang, W., Stevenson, D. K., Padula, A. M., Baiocchi, M., Burke, M., Cullen, M. R., & Shaw, G. M. (2020). Oil and gas production and spontaneous preterm birth in the San Joaquin Valley, CA: A case–control study. Environmental Epidemiology, 4(4), e099. https://doi.org/10.1097/EE9.0000000000000099 Guarnieri, M., & Balmes, J. R. (2014). Outdoor air pollution and asthma. The Lancet, 383(9928), 1581–1592. https://doi.org/10.1016/S0140-6736(14)60617-6 Ha, S., Hu, H., Roussos-Ross, D., Haidong, K., Roth, J., & Xu, X. (2014). The effects of air pollution on adverse birth outcomes. Environmental Research, 134, 198–204. https://doi.org/10.1016/j.envres.2014.08.002
Helmig, D. (2020). Air quality impacts from oil and natural gas development in Colorado.

Elementa: Science of the Anthropocene, 8(4). https://doi.org/10.1525/elementa.398

21 Hill, A. B. (1965). The Environment and Disease: Association or Causation? Proceedings of the Royal Society of Medicine, 58(5), 295–300.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1898525/
Hill, E. L. (2018). Shale gas development and infant health: Evidence from Pennsylvania.

Journal of Health Economics, 61, 134–150. https://doi.org/10.1016/j.jhealeco.2018.07.004 Hill, L. L., Czolowski, E. D., DiGiulio, D., & Shonkoff, S. B. C. (2019). Temporal and spatial trends of conventional and unconventional oil and gas waste management in Pennsylvania, 1991–2017. Science of The Total Environment, 674, 623–636. https://doi.org/10.1016/j.scitotenv.2019.03.475
Holder, C., Hader, J., Avanasi, R., Hong, T., Carr, E., Mendez, B., Wignall, J., Glen, G., Guelden, B., Wei, Y., & CDPHE (Colorado Department of Public Health and Environment). (2019). Evaluating potential human health risks from modeled inhalation exposures to volatile organic compounds emitted from oil and gas operations. Journal of the Air & Waste Management Association, 69(12), 1503–1524. https://doi.org/10.1080/10962247.2019.1680459
Jackson, R. B., Vengosh, A., Carey, J. W., Davies, R. J., Darrah, T. H., O’Sullivan, F., & Pétron, G. (2014). The Environmental Costs and Benefits of Fracking. Annual Review of Environment and Resources, 39(1), 327–362. https://doi.org/10.1146/annurev-environ- 031113-144051
Janitz, A. E., Dao, H. D., Campbell, J. E., Stoner, J. A., & Peck, J. D. (2019). The association between natural gas well activity and specific congenital anomalies in Oklahoma, 1997- 2009. Environment International, 122, 381–388.
https://doi.org/10.1016/j.envint.2018.12.011
Johnston, J. E., Enebish, T., Eckel, S. P., Navarro, S., & Shamasunder, B. (2021).

Respiratory Health, Pulmonary Function and Local Engagement in Urban Communities Near Oil Development. Environmental Research, 111088.
https://doi.org/10.1016/j.envres.2021.111088
Kan, H., Heiss, G., Rose, K. M., Whitsel, E., Lurmann, F., & London, S. J. (2007). Traffic exposure and lung function in adults: The Atherosclerosis Risk in Communities study.

Thorax, 62(10), 873–879. https://doi.org/10.1136/thx.2006.073015 Koehler, K., Ellis, J. H., Casey, J. A., Manthos, D., Bandeen-Roche, K., Platt, R., & Schwartz, B. S. (2018). Exposure Assessment Using Secondary Data Sources in Unconventional Natural Gas Development and Health Studies. Environmental Science & Technology, 52(10), 6061–6069. https://doi.org/10.1021/acs.est.8b00507 Lauer, N. E., Warner, N. R., & Vengosh, A. (2018). Sources of Radium Accumulation in Stream Sediments near Disposal Sites in Pennsylvania: Implications for Disposal of Conventional Oil and Gas Wastewater. Environmental Science & Technology. https://doi.org/10.1021/acs.est.7b04952
Li, C., Yang, M., Zhu, Z., Sun, S., Zhang, Q., Cao, J., & Ding, R. (2020). Maternal exposure to air pollution and the risk of low birth weight: A meta-analysis of cohort studies.

Environmental Research, 190, 109970. https://doi.org/10.1016/j.envres.2020.109970

22 Li, L., Blomberg, A. J., Spengler, J. D., Coull, B. A., Schwartz, J. D., & Koutrakis, P. (2020).

Unconventional oil and gas development and ambient particle radioactivity. Nature Communications, 11(1), 5002. https://doi.org/10.1038/s41467-020-18226-w Liu, L., Johnson, H. L., Cousens, S., Perin, J., Scott, S., Lawn, J. E., Rudan, I., Campbell, H., Cibulskis, R., Li, M., Mathers, C., & Black, R. E. (2012). Global, regional, and national causes of child mortality: An updated systematic analysis for 2010 with time trends since 2000. The Lancet, 379(9832), 2151–2161. https://doi.org/10.1016/S0140-6736(12)60560- 1 Long, J. C. S., Birkholzer, J. T., & Feinstein, L. C. (2015). An Independent Scientific Assessment of Well Stimulation in California: Summary Report—An Examination of Hydraulic Fracturing and Acid Stimulations in the Oil and Gas Industry. https://ccst.us/wp- content/uploads/2015SB4summary.pdf
Lucas, R. M., & McMichael, A. J. (2005). Association or causation: Evaluating links between “environment and disease”. Bulletin of the World Health Organization, 83(10), 792–795. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2626424/
Ma, Z.-Q. (2016). Time Series Evaluation of Birth Defects in Areas with and without Unconventional Natural Gas Development. Journal of Epidemiology and Public Health Reviews, 1. https://doi.org/10.16966/2471-8211.107
McKenzie, L. M., Allshouse, W., & Daniels, S. (2019). Congenital heart defects and intensity of oil and gas well site activities in early pregnancy. Environment International, 132, 104949. https://doi.org/10.1016/j.envint.2019.104949
McKenzie, L. M., Blair, B. D., Hughes, J., Allshouse, W. B., Blake, N., Helmig, D., Milmoe, P., Halliday, H., Blake, D. R., & Adgate, J. L. (2018). Ambient Non-Methane Hydrocarbon Levels Along Colorado’s Northern Front Range: Acute and Chronic Health Risks.

Environmental Science & Technology. https://doi.org/10.1021/acs.est.7b05983 McKenzie, L. M., Guo, R., Witter, R. Z., Savitz, D. A., Newman, L. S., & Adgate, J. L. (2014).

Birth outcomes and maternal residential proximity to natural gas development in rural Colorado. Environmental Health Perspectives, 122(4), 412–417. https://doi.org/10.1289/ehp.1306722
Michanowicz, D. R., Lebel, E. D., Domen, J. K., Hill, L. L., Jaeger, J. M., Schiff, J. E., Krieger, E. M., Banan, Z., Nordgaard, C. L., & Shonkoff, S. B. C. (Forthcoming). Methane and Health-damaging Air Pollutants From the Oil and Gas Sector: Bridging 10 Years of Scientific Understanding. PSE Healthy Energy.

Moore, C. W., Zielinska, B., Pétron, G., & Jackson, R. B. (2014). Air Impacts of Increased Natural Gas Acquisition, Processing, and Use: A Critical Review. Environmental Science & Technology, 48(15), 8349–8359. https://doi.org/10.1021/es4053472 NIEHS (National Institute of Environmental Health Sciences). (2021). Air Pollution and Your Health. National Institute of Environmental Health Sciences. https://www.niehs.nih.gov/health/topics/agents/air-pollution/index.cfm Nieuwenhuijsen, M. J., Ristovska, G., & Dadvand, P. (2017). WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and

23 Adverse Birth Outcomes. International Journal of Environmental Research and Public Health, 14(10), E1252. https://doi.org/10.3390/ijerph14101252 Pavley (2013). Senate Bill No. 4 Oil and Gas: Well Stimulation. https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201320140SB4 Peng, L., Meyerhoefer, C., & Chou, S.-Y. (2018). The health implications of unconventional natural gas development in Pennsylvania. Health Economics, 27(6), 956–983. https://doi.org/10.1002/hec.3649
Rabinowitz, P. M., Slizovskiy, I. B., Lamers, V., Trufan, S. J., Holford, T. R., Dziura, J. D., Peduzzi, P. N., Kane, M. J., Reif, J. S., Weiss, T. R., & Stowe, M. H. (2015). Proximity to Natural Gas Wells and Reported Health Status: Results of a Household Survey in Washington County, Pennsylvania. Environmental Health Perspectives, 123(1), 21–26. https://doi.org/10.1289/ehp.1307732
Rasmussen, S. G., Ogburn, E. L., McCormack, M., Casey, J. A., Bandeen-Roche, K., Mercer, D. G., & Schwartz, B. S. (2016). Association Between Unconventional Natural Gas Development in the Marcellus Shale and Asthma Exacerbations. JAMA Internal Medicine, 176(9), 1334. https://doi.org/10.1001/jamainternmed.2016.2436 Schraufnagel, D. E., Balmes, J. R., De Matteis, S., Hoffman, B., Kim, W. J., Perez-Padilla, R., Rice, M., Sood, A., Vanker, A., & Wuebbles, D. J. (2019). Health Benefits of Air Pollution Reduction. Annals of the American Thoracic Society, 16(12), 1478–1487. https://doi.org/10.1513/AnnalsATS.201907-538CME
Seixas, N. S., & Checkoway, H. (1995). Methodology Series: Exposure Assessment in Industry Specific Retrospective Occupational Epidemiology Studies. Occupational and Environmental Medicine, 52(10), 625–633. https://www.jstor.org/stable/27730414 Shamasunder, B., Collier-Oxandale, A., Blickley, J., Sadd, J., Chan, M., Navarro, S., Hannigan, M., & Wong, N. J. (2018). Community-Based Health and Exposure Study around Urban Oil Developments in South Los Angeles. International Journal of Environmental Research and Public Health, 15(1). https://doi.org/10.3390/ijerph15010138 Shapiro, G. D., Fraser, W. D., Frasch, M. G., & Séguin, J. R. (2013). Psychosocial stress in pregnancy and preterm birth: Associations and mechanisms. Journal of Perinatal Medicine, 41(6), 631–645. https://doi.org/10.1515/jpm-2012-0295 Shonkoff, S. B. C., Maddalena, R. L., Hays, J., Stringfellow, W. T., Wettstein, Z. S., Harrison, R., Sandelin, W., & McKone, T. E. (2015). Chapter 6: Potential Impacts of Well Stimulation on Human Health in California (An Independent Scientific Assessment of Well Stimulation in California, Volume II: Generic and Potential Environmental Impacts of Well Stimulation Treatments). California Council on Science and Technology. https://ccst.us/wp-content/uploads/160708-sb4-vol-II-6-1.pdf Soriano, M. A., Siegel, H. G., Gutchess, K. M., Clark, C. J., Li, Y., Xiong, B., Plata, D. L., Deziel, N. C., & Saiers, J. E. (2020). Evaluating Domestic Well Vulnerability to Contamination From Unconventional Oil and Gas Development Sites. Water Resources Research, n/a(n/a), e2020WR028005. https://doi.org/10.1029/2020WR028005

24 Stacy, S. L., Brink, L. L., Larkin, J. C., Sadovsky, Y., Goldstein, B. D., Pitt, B. R., & Talbott, E.

O. (2015). Perinatal Outcomes and Unconventional Natural Gas Operations in Southwest Pennsylvania. PLOS ONE, 10(6), e0126425.
https://doi.org/10.1371/journal.pone.0126425
Stringfellow, W. T., Camarillo, M. K., Domen, J. K., & Shonkoff, S. B. (2017). Comparison of chemical-use between hydraulic fracturing, acidizing, and routine oil and gas development. PloS One, 12(4), e0175344.

Swinburn, T. K., Hammer, M. S., & Neitzel, R. L. (2015). Valuing Quiet: An Economic Assessment of U.S. Environmental Noise as a Cardiovascular Health Hazard. American Journal of Preventive Medicine, 49(3), 345–353.
https://doi.org/10.1016/j.amepre.2015.02.016
Tang, I. W., Langlois, P. H., & Vieira, V. M. (2021). Birth defects and unconventional natural gas developments in Texas, 1999–2011. Environmental Research, 194, 110511. https://doi.org/10.1016/j.envres.2020.110511
Tran, K. V., Casey, J. A., Cushing, L. J., & Morello-Frosch, R. (2020). Residential Proximity to Oil and Gas Development and Birth Outcomes in California: A Retrospective Cohort Study of 2006–2015 Births. Environmental Health Perspectives, 128(6), 067001. https://doi.org/10.1289/EHP5842
Tran, K. V., Casey, J. A., Cushing, L. J., & Morello-Frosch, R. (Forthcoming). Residential proximity to hydraulically fractured oil and gas wells and adverse birth outcomes in urban and rural communities in California (2006-2015). Environmental Epidemiology.

US EPA. (2019). Integrated Science Assessment (ISA) for Particulate Matter (Final Report, Dec 2019) [Reports & Assessments].
https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=347534 US EPA. (2020). Integrated Science Assessment (ISA) for Ozone and Related Photochemical Oxidants (Final Report, Apr 2020) [Reports & Assessments]. https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=348522 US EPA. (2021, April). BenMAP: Environmental Benefits Mapping and Analysis Program – Community Edition. User’s Manual. https://www.epa.gov/sites/default/files/2015- 04/documents/benmap-ce_user_manual_march_2015.pdf
Vogel, J. P., Chawanpaiboon, S., Moller, A.-B., Watananirun, K., Bonet, M., & Lumbiganon, P. (2018). The global epidemiology of preterm birth. Best Practice & Research Clinical Obstetrics & Gynaecology, 52, 3–12. https://doi.org/10.1016/j.bpobgyn.2018.04.003 Walker Whitworth, K., Kaye Marshall, A., & Symanski, E. (2018). Drilling and Production Activity Related to Unconventional Gas Development and Severity of Preterm Birth.

Environmental Health Perspectives, 126(3), 037006. https://doi.org/10.1289/EHP2622 Whitworth, K. W., Marshall, A. K., & Symanski, E. (2017). Maternal residential proximity to unconventional gas development and perinatal outcomes among a diverse urban population in Texas. PLOS ONE, 12(7), e0180966.
https://doi.org/10.1371/journal.pone.0180966

25 Willis, M. D., Hill, E. L., Boslett, A., Kile, M. L., Carozza, S. E., & Hystad, P. (2021).

Associations between Residential Proximity to Oil and Gas Drilling and Term Birth Weight and Small-for-Gestational-Age Infants in Texas: A Difference-in-Differences Analysis.

Environmental Health Perspectives, 129(7), 077002. https://doi.org/10.1289/EHP7678 Willis, M., Hystad, P., Denham, A., & Hill, E. (2020). Natural gas development, flaring practices and paediatric asthma hospitalizations in Texas. International Journal of Epidemiology. https://doi.org/10.1093/ije/dyaa115
Willis, M. D., Jusko, T. A., Halterman, J. S., & Hill, E. L. (2018). Unconventional natural gas development and pediatric asthma hospitalizations in Pennsylvania. Environmental Research, 166, 402–408. https://doi.org/10.1016/j.envres.2018.06.022 Zammerilli, A., Murray, R., Davis, T., & Littlefield, J. (2014). Environmental impacts of unconventional natural gas development and production. DOE/NETL-2014/1651. U.S.

DOE.
https://wedocs.unep.org/bitstream/handle/20.500.11822/17944/netl_environmental.pdf?se quence=1&isAllowed=y

Aquatic Toxicology 230 (2021) 105714
Available online 7 December 2020
0166-445X/© 2020 Elsevier B.V. All rights reserved.

Hepatobiliary PAHs and prevalence of pathological changes in Red Snapper Erin L. Pulster
a ,
* , Susan Fogelson
b
, Brigid E. Carr
a
, Justin Mrowicki
a
, Steven A. Murawski
a
a
University of South Florida, College of Marine Science, St. Petersburg, FL, USA b
Fishhead Labs, LLC, Stuart, FL, USA

ARTICLE INFO
Keywords:

Liver
Pathology
DWH
Fish
Polycyclic aromatic hydrocarbons
Gulf of Mexico
Microscopic hepatic changes

ABSTRACT
Red Snapper (Lutjanus campechanus) were collected throughout the Gulf of Mexico (GoM) from 2011 to 2017 and analyzed for biliary ( n = 496) fluorescent aromatic compounds (FACs), hepatic ( n = 297) polycyclic aromatic hydrocarbons (PAHs) and microscopic hepatobiliary changes (MHC, n = 152). Gross and histological evaluations were conducted with liver tissues to identify and characterize pathological changes. This is the first report to interrelate hepatobiliary PAH concentrations and MHCs in Red Snapper. Hepatic PAHs measured in GoM Red Snapper ranged from 192 to 8530 ng g
1
w.w. and biliary FACs ranged from 480 to 1,100,000 ng FAC g 1
bile.

Biliary FACs in Red Snapper collected along the west Florida Shelf and north central region declined after 2011 and were relatively stable until a sharp increase was noted in 2017. Increases in the PAH exposures are likely due to a number of sources including leaking infrastructure, annual spills, riverine input and the resuspension of contaminated sediments. In contrast, hepatic PAH concentrations were relatively stable indicating Red Snapper are able to maintain metabolic clearance however this energetic cost may be manifesting as microscopic hepatic changes (MHCs). Virtually all (99 %) of the evaluated Red Snapper had one to nine MHCs with an average of five coinciding changes in an individual fish. The observed changes were broadly classified as inflammatory re - sponses, metabolic responses, degenerative lesions, nonneoplastic proliferation and neoplastic lesions. Biliary FACs were associated with parasitic infection and intracellular breakdown product accumulation such as intra- macrophage hemosiderin, lipofuscin and ceroid laden prevalence. Whereas, hepatic PAHs were associated with increased myxozoan plasmodia prevalence. This study evaluates relationships between hepatobiliary PAH con - centrations and biometrics, somatic indices, condition factors and microscopic hepatic changes in Red Snapper located in the north central GoM. Together, these results may be signaling increased disease progression in Gulf of Mexico Red Snapper more than likely resulting from chronic environmental stressors including elevated PAH exposures and concentrations.

1. Introduction
Red Snapper (Lutjanus campechanus) are the dominant species of the mid-shelf reef complex distributed along the continental shelf throughout the United States (U.S.) and Mexican waters of the Gulf of Mexico (GoM) and the southeast U.S. Atlantic coast. This is a long-lived demersal species ( ≤ 50 years) that significantly supports both recrea - tional and commercial fisheries. The latest stock assessment of GoM Red Snapper indicates the stock is no longer overfished but continues to rebuild from the severely overfished and depleted conditions experi - enced over the past few decades. Due to the ecological and economic importance of Red Snapper in the GoM, this species was the focus of a number of impact-related studies following the Deepwater Horizon (DWH) oil spill in 2010. Post-DWH, Red Snapper exhibited shifts in diet and trophic level ( Tarnecki and Patterson, 2015 ), declines in growth ( Herdter et al., 2017 ) and abundance ( Pulster et al., 2020c ), and an increased frequency of epidermal lesions ( Murawski et al., 2014 ).

The most toxic component of oil, polycyclic aromatic hydrocarbons (PAHs), are of great interest due to their association with a wide range of deleterious health effects in fish ( Collier et al., 2013 ; Pulster et al., 2020c ). Common mechanisms of toxicity include metabolic activation of PAHs generating reactive intermediates or metabolites, the activation of the aryl hydrocarbon receptor (AHR) and AHR-dependent alterations in gene expression. Persistent AHR activation by planar xenobiotics, including many PAHs, potentially mediates various toxicological events in tissues as well as vascular and immune systems ( Curtis et al., 2011 ).

Toxic chemicals and other xenobiotics are primarily metabolized in the liver of vertebrates. The morphological features of a fish liver have the * Corresponding author at: University of South Florida, College of Marine Science, 140 7 th
Avenue South, St. Petersburg, Florida, 33701, USA.

E-mail address: [email protected] (E.L. Pulster).

Contents lists available at ScienceDirect
Aquatic Toxicology
journal homepag e: www.else vier.com/loc ate/aqtox
https://doi.org/10.1016/j.aquatox.2020.105714
Received 8 June 2020; Received in revised form 24 November 2020; Accepted 2 December 2020 Aquatic Toxicology 230 (2021) 105714
2 same general circulatory (i.e., hepatic arterioles, portal veins, hepatic veins) and biliary conduits as their mammalian counterpart ( Wolf and Wolfe, 2005 ). Most fish species have a single-lobed liver with two main circulatory regions of the vasculature. However, there are subtle inter - species differences in the sinusoidal structure and considerable inter - species differences in the length and position of the ducts making up the biliary system ( Gingerich, 1982 ). The conformation of fish hepatocytes are often described as double-layered cords representing blind-ended, anastomosing and branching tubules with a lack of apparent organiza - tion. Furthermore, histological findings tend to reveal a more vacuolated (higher glycogen and/or lipid content) compared to mammals ( Gin - gerich, 1982 ; Wolf and Wolfe, 2005 ). All fish have a resident population of hepatic macrophages as part of the reticuloendothelial system. While they can be histologically unapparent in healthy fish, unique species anatomic variation or activation of the reticuloendothelial system can present as individualized or small nests of resident pigmented macro - phage aggregates (PMAs). Nonresident perisinusoidal macrophages and hepatic stellate cells (Ito cells, perisinusoidal cells, fat-storing cells, lipocytes) can also be seen in the normal hepatic structure.

Physiologically, the same basic metabolic functions of the liver are also similar between fish and mammals, including energy metabolism and storage, storage of nutrients, synthesis of enzymes and other co - factors, bile formation and excretion (food digestion), hormone pro - duction (e.g., vitellogenin), and xenobiotic metabolism ( Wolf and Wheeler, 2018 ; Wolf and Wolfe, 2005 ). Fish have a well-developed biotransformation system capable of converting xenobiotics (e.g., PAHs) into water-soluble forms that can be easily eliminated but they lack a highly developed deoxyribonucleic acid (DNA) repair system to combat DNA damage caused by xenobiotics prior to elimination ( Var - anasi et al., 1987 ). For instance, the PAH metabolic intermediates of benzo[ a ]pyrene (e.g., 7,8-oxide benzo[ a ]pyrene and 7,8-dihydrodiol benzo[ a ]pyrene) have demonstrated genotoxic and carcinogenic prop - erties by forming DNA adducts which induce mutations leading to cancer. The liver is also a common target for both cytotoxicity and tumorigenesis in part due to biotransformation reactions that can enhance toxicity and carcinogenicity of metabolites ( Hinton et al., 2001 ; Wolf and Wheeler, 2018 ). Additionally, the enterohepatic cycling mechanism in fish prolongs the removal of certain compounds and thus creates the potential for increased organ accumulation and toxicity ( Gingerich, 1982 ). Resultingly, biliary PAH metabolites in fish serve as a sensitive indicator of recent oil exposure (e.g., days to weeks), whereas, PAHs in tissues serve as an indicator of long-term (months to years) chronic exposure.

Histopathological alterations or microscopic hepatic changes (MHCs) at both the molecular and cellular level of an organism have been documented in various organs (e.g., liver, kidney, gill) of fish following oil exposures ( Brown-Peterson et al., 2015 ; Katsumiti et al., 2008 ; Marty et al., 2003 ; Whitehead et al., 2012 ). Examples of MHCs observed in fish liver following exposures to oil include hepatic intra - vascular congestion, lipid changes, necrosis, cystic degeneration, foci of cellular alteration (FCA), and neoplasia ( Brown-Peterson et al., 2015 ; Oliva et al., 2013 ; Reddy et al., 1999 ). These MHCs, as well as others described in the literature, provide definitive biological end-points of historical exposures and has led to their increased usage as indicators of environmental contamination ( Oliva et al., 2013 ).

Previous studies have reported chronic and increasing hepatobiliary PAH concentrations in GoM fishes ( Pulster et al., 2020d ), including ecologically and economically important demersal species including Golden Tilefish Lopholatilus chamaeleonticeps ( Snyder, 2020 ), hakes Urophycis spp.( Struch et al., 2019 ), and ten species of grouper ( Pulster et al., 2020a ). This documented increase in PAH exposure has contrib - uted to growing concern regarding the potential adverse health effects in fish as well as the continued degradation of the GoM ecosystem as a whole. In order to help understand the impacts of chronic PAH expo - sures on Red Snapper, we developed this study with the objectives to (1) survey hepatobiliary PAH concentrations throughout their range in the GoM; (2) evaluate temporal trends in PAH exposures in the DWH impacted area; and (3) correlate hepatic PAH concentrations with hepatobiliary health of Red Snapper in the GoM.

2. Materials and methods

2.1. Field collections
Red Snapper ( n = 566) were collected from 72 locations during May through September of 2011 2017. Fishing was conducted using demersal longline sampling gear along transects throughout the GoM with transect locations divided into six regions of the GoM designated as the north central (NC), West Florida Shelf (WFS), Yucatan Shelf (YS), Bay of Campeche (BC), southwest (SW), and northwest (NW) regions ( Fig. 1 ). Biometrics were recorded upon retrieval (Table A1), bile and hepatic tissues were collected from a subsample of Red Snapper for PAH analyses and histopathology. The demersal longline sampling design and protocols have been previously described in detail ( Murawski et al., 2018 ; Pulster et al., 2020a ) and brief sampling specifics can be found in the Appendix.

2.2. Biliary and hepatic PAH analysis
Untreated bile samples (3 μ L) from Red Snapper ( n = 496) were analyzed for naphthalene, phenanthrene and benzo[ a ]pyrene metabo - lite equivalents (also known as fluorescent aromatic compounds, FACs) using high performance liquid chromatography fluorescence detection (HPLC-F). Biliary FACs are expressed as ng FAC g
1
bile and reported as
the sum of all three PAH metabolite equivalents (TFAC) rounded to two significant figures. Homogenized hepatic tissues ( n = 297) were pre - pared for analysis using QuEChERS extraction and clean-up methodol - ogy (Bond Elut, Agilent Technologies, Santa Clara, CA, USA). Target PAHs in liver tissues were confirmed and quantified using a gas chro - matograph (GC, Agilent 7890B) coupled to a triple quadrupole mass spectrometer (MS/MS, Agilent 7010) operating in multiple reaction monitoring (MRM) mode. Hepatic PAHs are reported as the sum of 46 PAH analytes (T
46 PAH) rounded to three significant figures expressed as ng g
1
(w.w.). The 2 – 3 ring and 4 – 6 ring PAHs and alkylated homologs were summed as low molecular weight (LMW) and high molecular weight (HMW) PAHs, respectively. Further details on specific methods, instrument parameters and quality assurance protocols are provided in the Appendix.

2.3. Health proxies
Providing Red Snapper hepatic samples had sufficient mass for both PAH analysis and lipid extractions, lipid content was measured using a modified Folch method described in Pulster et al. (2020b) . Fulton ’ s Condition Factor [K = (total weight (g) / standard length (cm 3 ) x 100]
and hepatic (HSI), gastrointestinal (GaSI) and gonadal (GSI) somatic indices [SI = organ weight (g) / total weight (g)] were calculated for each individual fish.

2.4. Histopathology
Gross and histological evaluations were performed on 152 hepatic samples from 17 locations in the NC ( n = 91), NW ( n = 57), and SW ( n = 4) regions of the GoM collected 2012 2017. In the field, three repre - sentative segments were taken from the middle and distal portions of the liver lobes from each individual. Microscopic examination was per - formed using a Nikon Eclipse 80i (Nikon, Minato, Tokyo, Japan) and photomicrographs were taken with an Accu-scope Excelis HD (Com - mack, NY). Specifics on the tissue preparation and analysis are located in the Appendix.

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
3 2.5. Statistical analyses
All statistical analyses were performed using JMP Pro Version 14.3 (SAS Institute) and MATLAB R2019b (Update 3, MathWorks) with the Fathom Toolbox. The Appendix provides more in-depth detail on the methodology used for data analyses.

3. Results

3.1. Associations between PAH concentrations and biometrics Hepatic PAHs and biliary FACs in Red Snapper ranged from 192 to 8530 ng g
1
w.w. and 480 to 1,100,000 ng FAC g
1
bile, respectively
( Table 1 , Fig. 2 ). No significant relationships were found between biliary FACs and liver lipid content ( Table 2 ). Weak (r = 0.30 0.50) to Fig. 1. Collection locations for Red Snapper ( n = 566) used in this study. Dashed lines indicate the delineations between regions.

Table 1
Mean ( ± standard deviation) and total biliary fluorescent aromatic compounds (TFAC) and hepatic polycyclic aromatic hydrocarbon (T 46 PAH) concentrations
measured in Red Snapper collected in the Gulf of Mexico, 2011-2017.

Biliary FACs (ng FAC g
1
bile) Hepatic PAHs (ng g
1
w.w.)
Region Year Nap Phn B[ a ]P TFACs LMW HMW T
46 PAH
Bay of Campeche (BC)
2015 22,000 ± 11,000 N/A 170 ± 80 22,000 ± 11,000 1870 ± 1190 19.1 ± 6.38 1890 ± 1190 2016 44,000 ± 49,000 11,000 ± 9000 110 ± 70 55,000 ± 58,000 1022 ± 546 9.84 ± 3.08 1031 ± 546 2015 2016 28,000 ± 26,000 11,000 ± 9000 160 ± 80 31,000 ± 32,000 1480 ± 1012 14.9 ± 6.91 1500 ± 1016 North Central (NC)
2011 120,000 ± 78,000 21,000 ± 12,000 280 ± 140 140,000 ± 90,000 1203 ± 977 9.39 ± 7.09 1210 ± 979 2012 60,000 ± 27,000 13,000 ± 4300 220 ± 150 74,000 ± 30,000 1240 ± 701 6.69 ± 2.50 1250 ± 701 2013 56,000 ± 26,000 13,000 ± 6000 450 ± 330 69,000 ± 32,000 2640 ± 2600 9.21 ± 5.74 2650 ± 2600 2014 76,000 ± 70,000 N/A 440 ± 270 76,000 ± 70,000 955 ± 375 5.23 ± 2.13 960 ± 375 2015 65,000 ± 61,000 N/A 340 ± 270 65,000 ± 61,000 1540 ± 1110 9.99 ± 5.18 1550 ± 1110 2017 140,000 ± 130,000 26,000 ± 17,000 210 ± 120 170,000 ± 150,000 1094 ± 300 4.54 ± 1.59 1099 ± 301 2011 2017 80,000 ± 73,000 17,000 ± 11,000 350 ± 270 90,000 ± 81,000 1480 ± 1290 8.21 ± 5.12 1490 ± 1290 Northwest (NW) 2016 110,000 ± 85,000 19,000 ± 15,000 350 ± 530 130,000 ± 92,000 2330 ± 1460 10.2 ± 5.97 2340 ± 1470 2017 220,000 ± 170,000 40,000 ± 22,000 240 ± 190 250,000 ± 190,000 N/A N/A N/A 2016 2017 130,000 ± 120,000 23,000 ± 18,000 330 ± 480 150,000 ± 130,000 2330 ± 1460 10.2 ± 5.97 2340 ± 1470 Southwest (SW) 2016 130,000 ± 120,000 40,000 ± 44,000 170 ± 80 170,000 ± 150,000 1310 ± 988 8.43 ± 5.69 1320 ± 993 West Florida Shelf (WFS)
2011 N/A N/A N/A N/A 724 ± 389 14.7 ± 6.91 738 ± 390
2013 31,000 ± 24,000 5000 ± 2800 110 ± 90 36,000 ± 26,000 N/A N/A N/A 2014 31,000 ± 22,000 8000 ± 5000 130 ± 70 37,000 ± 24,000 403 ± 113 6.44 ± 1.89 409 ± 113 2015 10,000 ± 6900 N/A 100 ± 130 11,000 ± 7000 727 ± 397 12.1 ± 3.33 740 ± 397 2017 210,000 ± 91,000 49,000 ± 23,000 250 ± 180 260,000 ± 110,000 N/A N/A N/A 2011 2017 98,000 ± 110,000 27,000 ± 27,000 170 ± 150 120,000 ± 130,000 648 ± 367 12.5 ± 6.71 661 ± 369 Yucatan Shelf (YS) 2015 38,000 ± 43,000 N/A 74 ± 90 38,000 ± 43,000 659 ± 359 6.49 ± 5.15 666 ± 353 FAC = fluorescent aromatic compounds; Nap = Naphthalene, Phn = Phenanthrene, B[ a ]P = benzo[ a ]pyrene, TFAC = sum of Nap + Phn + B[ a ]P FACs, LMW = low molecular weight PAHs, HMW = high molecular weight PAHs, T 46 PAH = total of 46 PAHs.

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
4 moderate (r = 0.50 0.70) negative associations were found between hepatic lipids and both wet weight (r = -0.32, p < 0.0001) and lipid normalized (r = -0.49, p < 0.0001) PAH concentrations, therefore only wet weight concentrations are reported. Weak to very weak (r < 0.30) relationships were found between hepatobiliary PAHs and biometrics, somatic indices and condition factor ( Table 2 ). There was also a weak positive relationship detected between biliary and hepatic PAHs con - centrations (r = 0.21, p = 0.003). Weak to strong positive associations were detected between the HSI and GSI (r = 0.484, p = 0.007), HSI and GaSI (r = 0.904, p = 0.001) and GSI and GaSI (r = 0.416, p = 0.006) in Red Snapper collected Gulf-wide.

In the NW region, the total weight of female Red Snapper was significantly larger than males ( p = 0.03). There were no other signifi - cant differences observed between sex and length ( p = 0.55), weight ( p = 0.29), or condition factor ( p = 0.43) for Red Snapper collected Gulf- wide or within a particular region (all years combined). The total lengths (TL) of Red Snapper used in this study ranged between 31 and 101 cm (63 ± 13 cm) (Table A1). Based on the TL in this dataset, the estimated age range is between 2 and 25 years or older, with a mean of approximately 6 years old ( Herdter et al., 2017 ; Patterson et al., 2001 ).

Strong positive relationships were observed between standard length and total weight (r = 0.912, p = 0.001).

There were no differences in hepatobiliary PAH concentrations be - tween male and female Red Snapper even though mean ( ± standard deviation) hepatic lipid content was significantly higher in males (0.49 ± 0.24 g, n = 169) compared to females (0.40 ± 0.18 g, n = 129, p =

0.003) for all Gulf-wide data combined over years and regions.

Regionally, sex differences in hepatobiliary concentrations were only detected in the NC region, with males having significantly higher biliary FAC (104,000 ± 98,000 ng FAC g
1
, p = 0.021) and hepatic (1730 ±
Fig. 2. Mean biliary total fluorescent aromatic compounds (TFACs, A) and hepatic total petroleum hydrocarbons (TPAHs, B) in Gulf of Mexico Red Snapper, 2011- 2017. The dashed lines delineate between six regions designated as the West Florida Shelf (WFS), north central (NC), northwest (NW), southwest (SW), Bay of Campeche (BC), and Yucat ´an Shelf (YS).

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
5 1580 ng g
1
w.w., p = 0.009) PAH concentrations compared to females (75,000 ± 53,000 ng FAC g
1
; 1160 ± 665 ng g
1
w.w.).

3.2. Gulf-wide spatial trends in biometrics and PAH concentrations For all years combined, regional differences using permanovas were detected in the standard lengths ( F = 6.18, p = 0.001), total fish weight ( F = 2.85, p = 0.019), HSI ( F = 6.13, p = 0.001), condition factors ( F =

17.70, p = 0.001) and hepatic lipid content ( F = 36.42, p = 0.001).

Pairwise permanovas determined the lipid content ( p = 0.015) of NC Red Snapper were significantly larger than those along the WFS. The HSI ( p = 0.015) and GSI ( p = 0.015) of Red Snapper collected in the NC region were significantly larger than those collected in the NW region.

Additionally, Red Snapper in the NC region had a lower condition factor compared to the NW ( p = 0.015), WFS ( p = 0.015), SW ( p = 0.045) and BC ( p = 0.015) regions. Red Snapper collected along the WFS had significantly larger HSI than those collected in the NW ( p = 0.015) and SW ( p = 0.015) regions as well as significantly larger GSI than the NW ( p = 0.03).

In 2013 ( p = 0.002), 2014 ( p = 0.012), and 2015 ( p = 0.018) mean biliary FACs measured in Red Snapper from the NC region were signif - icantly higher than Red Snapper collected along the WFS. In 2015, mean biliary and hepatic concentrations in the NC region were also signifi - cantly higher than fish collected in the BC ( p = 0.024) and WFS ( p =

0.042) regions, respectively. No other spatial trends were detected in hepatobiliary PAH concentrations.

3.3. Temporal trends in the North Central Region
The current study expands upon an earlier study documenting the decline in biliary FACs in Red Snapper collected in the NC region of the GoM between 2011 and 2013 ( Snyder et al., 2015 ) by providing an extended time series for biliary FAC concentrations (2014 – 2017) and including PAH levels for hepatic tissues (2011 2017). Temporal dif - ferences detected in the biometrics of Red Snapper collected in the NC region are summarized in Table A1. The HSIs of Red Snapper collected in the NC region during 2014 were significantly higher than Red Snapper collected in the same region during 2011 ( p = 0.015), 2012 ( p = 0.015), 2013 ( p = 0.03), 2015 ( p = 0.015), and 2017 ( p = 0.015). The 2012 GSIs were significantly lower than in 2014 ( p = 0.015) and 2017 ( p = 0.015).

There were no significant differences observed between years for the GaSIs in the NC region. In general, all somatic indices for Red Snapper in the NC region increased between 2011 and 2014 (HSIs: 814 %, GaSIs:
190 %, GSIs: 292 %) followed by decreases by 2017 (HSIs: -88 %, GaSIs:
-68 %, GSIs: -59 %). Condition factors in 2017 were significantly higher than 2011 ( p = 0.015), 2012 ( p = 0.03), 2013 ( p = 0.015) and 2015 ( p =

0.03) and exhibited an overall 43 % increase between 2011 and 2017.

Mean biliary FACs for NC Red Snapper were significantly higher in 2011 compared to 2012 ( p = 0.015), 2013 ( p = 0.015), 2014 ( p = 0.045) and 2015 ( p = 0.015; Table 1 ). The 2017 mean biliary FAC concentra - tion was significantly higher than 2012 ( p = 0.015). In general, mean biliary FACs in the NC region declined 51 % between 2011 and 2013, and remained relatively stable until 2015, when a sharp increase (162 %) in concentrations was observed in 2017.

Although there were no significant temporal trends observed in he - patic PAH concentrations in Red Snapper collected in the NC region of the GoM ( Table 1 ), annual differences were detected in the hepatic profiles using discriminant analysis (Hotelling-Lawley p < 0.0001, Fig. 3 ). Approximately 82 % of the variability in the profiles is explained by year as a grouping variable for the factor 1 ( χ
2 (3815) = 556, p <

0.0001) and 2 ( χ
2 (2974) = 540, p < 0.0001) loadings, consisting of
primarily the LMW and HMW compounds, respectively ( Fig. 3 and A1).

3.4. Temporal trends on the West Florida shelf
There were no significant sex-related differences in standard lengths, total weight, organ weights, condition factor, hepatic lipid content, biliary FACs or hepatic PAHs in the WFS region for all years combined or within a particular year. Mean biliary FACs were significantly higher in 2017 compared to 2013 ( p = 0.006), 2014 ( p = 0.006) and 2015 ( p =

0.006) ( Table 1 ). Mean hepatic PAH concentrations were significantly higher in 2011 compared to 2014 ( p = 0.001) ( Table 1 ). Overall, mean biliary FACs in Red Snapper along the WFS increased 625 % between 2013 and 2017 whereas hepatic PAH concentrations decreased 45 % between 2011 and 2014, followed by an 81 % increase by 2015.

There were no significant annual differences in the HSIs or GSIs in Red Snapper collected along the WFS, however the GaSIs increased 64 % between 2011 and 2014 ( p = 0.01) and then decreased 31 % by 2015 ( p = 0.21). Condition factors of Red Snapper along the WFS declined 22 % between 2011 and 2013 ( p = 0.04) and then increased 36 % by 2017 ( p = 0.010).

3.5. Gross evaluation
Red Snapper ( n = 152) collected from three regions of the GoM (NC, Table 2
Prevalence (%) of microscopic hepatic changes (MHC) in Red Snapper collected by region (sample size in parentheses) in the Gulf of Mexico (GoM). Regions include the north central (NC), northwest (NW) and southwest (SW) GoM. The region with the highest prevalence for each MHC is bolded.

Microscopic Hepatic
Change
NC ( n =
91)
NW ( n =
57)
SW ( n =
4)
Gulf-wide ( n
= 152)
Lymphocytic inflammation 60 58 0 58
Parasites/nematodes/
metazoans
58 54 25 55
Pigmented macrophage
aggregates
57 60 25 57
Granuloma(s) 57 51 50 55
Cholangiofibrosis 47 68 25 55
Lipid-type vacuolar change 46 32 0 39
Glycogen-type vacuolar
change
35 39 50 37
Edema 34 0 0 20
Myxozoan plasmodia 12 21 0 13
Bile duct hyperplasia 10 2 0 7
Hepatocellular atrophy 9 4 0 7
Hepatocellular necrosis 8 7 0 7
Ito cell hyperplasia 1 9 0 4
Hyperplastic nodule 1 0 0 1
Foci of cellular alteration
(FCA)
0 2 0 1
Fig. 3. Discriminant analysis plot illustrating distinct clustering of PAH composition profiles by year in liver tissues of Red Snapper located in north central Gulf of Mexico. Ellipse indicate 90 % data coverage for each year.

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
6 NW, SW) between 2012 and 2017 were evaluated for gross and histo - logical changes in their hepatic tissues. Gross evaluation of Red Snapper hepatic tissues ( n = 152) revealed minimal morphologic variation. In general, hepatic tissues were comprised of homogenous soft tissue with a smooth capsular surface and no lobular delineation. Interspersed throughout the soft tissue were blood vessels as well as biliary ducts.

Blood vessels were often prominent and contained dark brown blood clots. The parenchyma ranged from light tan to dark tan and rare sam - ples had pinpoint to 1 mm dark brown or copper foci scattered in the hepatic capsule or throughout the parenchyma.

3.6. Histological evaluation
Histological evaluation of the hepatic tissues revealed a spectrum of hepatocellular morphologic variation in association with vacuolar change ranging from mild to severe and only one fish had a liver with no microscopic hepatic changes (MHCs). The normal microscopic anatomy of the liver is described in more detail in the Appendix. Microscopic hepatic changes were noted in 99 % of the fish evaluated ( Table 2 ). Each of the fish identified with changes had at least one MHC and as many as nine were noted for an individual fish. The 15 types of MHCs observed in GoM Red Snapper can be broadly categorized into: (1) inflammatory responses (granulomas, chronic inflammation, lymphocytic inflamma - tion, PMAs, fibrosis, edema, parasites); (2) metabolic responses (hepatocellular vacuolization, atrophy); (3) degenerative responses (necrosis); (4) nonneoplastic proliferative changes (FCAs, bile duct hy - perplasia, Ito cell hyperplasia); and (5) neoplastic lesions (hyperplastic nodule). A single occurrence of hepatocellular hyperplasia was noted in the NC region ( Fig. 4 a).

Lymphocytic inflammation was commonly observed in the peri - pancreatic connective tissue, adventitial tissue of the bile ducts, wall of vessels, or randomly distributed throughout the hepatic cords ( Fig. 4 b).

There was a moderate correlation between lymphocytic inflammation and parasites (R
2 = 0.49, p = 0.0190). Fifty-eight percent of the Red
Snapper identified with lymphocytic inflammation ( n = 88) had con - current granulomas containing organisms that were interpreted as larval nematodes ( Fig. 4 c). Granulomas were characterized by a center of necrotic cellular debris that variably contained multinucleated nema - tode larvae surrounded by concentric layers of flattened macrophages or a thin layer of fibrous connective tissue ( Fig. 4 c-d). Rare sub adult to adult nematodes were observed within the connective tissues adjacent to the pancreas or hepatic serosa (Figure A3).

Intra-biliary duct myxozoan plasmodia were observed in 13 % of the Red Snapper (Figure A3). Plasmodia were serpiginous, filled with floc - culent eosinophilic material and had various stages of developing myxospores. Parasitic plasmodia were observed in samples from the NC and NW regions. There was moderate cholangiofibrosis (bile duct fibrosis) in the affected fish ( Fig. 4 d). Cholangiofibrosis was variably Fig. 4. Photomicrograph of (A) hepatocellular
nodular hyperplasia (arrow) (H & E stain, 40X
magnification, 100 μ m scale bar); (B) focus of
lymphocytic inflammation (arrow) (H & E stain,
20X magnification, 25 μ m scale bar); (C) peri -
pancreatic multinucleated nematode larvae
encysted in fibrous connective tissue at the pe -
riphery of a peripancreatic granuloma (arrow,
H & E stain, 20X magnification); (D) bile duct
necrosis with adventitial lymphocytic inflam -
mation (arrow) and moderate peribiliary
fibrosis (cholangiofibrosis, H & E stain, 100X
magnification); (E) pigmented macrophage ag -
gregates in the peripancreatic tissue as well as
pigment within hepatocytes staining positive
for hemosiderin (arrow, Perl ’ s Iron stain, 100X
magnification); (F) pigment within macrophage
aggregates staining positive for PAS (arrow,
PAS stain, 100X magnification; (G) severe
glycogen type vacuolar change (H & E stain,
100X magnification); and (H) severe lipid type
vacuolar change (H & E stain, 100X magnifica -
tion). Scale bar for plates C-H is 50 μ m; H & E:

Hematoxylin and Eosin stain.

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
7 observed and severity often correlated with inflammation of the ducts or intraluminal organisms and was well-delineated by Masson ’ s trichrome stain as bright blue fibers.

Pigmented macrophage aggregates (PMA) contained foamy baso - philic cytoplasm pigmented with intracellular lipid droplets or hemo - siderin/lipofuscin/ceroid ( Fig. 4 e). Pigment within some macrophages stained positive with Perl ’ s Iron and PAS stain for hemosiderin ( Fig. 4 e- f) while for others the iron staining was negative (Figure A4). Sudan Black IV and AFB of the Perl ’ s negative aggregates stains were both negative indicating for some samples the intracellular pigments were not lipofuscin, lipid, triglycerides, or lipoprotein (Figure A5 6).

Intra-hepatocellular cytoplasmic vacuolization consistent with lipid (39 %) and glycogen (37 %) type vacuolar change were also prevalent in Red Snapper Gulf-wide and ranged from minimal to severe ( Fig. 4 g).

Lipid-type vacuolar change was characterized by intracytoplasmic, discrete round micro or macrovesicles that had distinct borders.

Glycogen-type vacuolar change was characterized by mild cellular swelling and diffuse replacement of the normal eosinophilic granular hepatocytic cytoplasm by wispy, vacuolated cytoplasm ( Fig. 4 h).

Hepatocellular atrophy was only observed in Red Snapper collected in the SW (2016) and NC (2017) regions. Atrophic hepatocytes were decreased in size due to shrunken cytoplasmic contents (Figure A7).

Atrophied hepatocytes often had deeply basophilic granular cytoplasm.

Due to cytoplasmic loss, nuclei appeared closer in proximity. Most of the samples that had atrophy were considered minimal to mild except for one fish collected near the Mississippi delta that was moderate in severity.

3.7. Biometric associations with MHCs
While lipid-type vacuolar change had a moderate positive relation - ship with liver lipid content (R
2 = 0.52, p = 0.012), lipid-type vacuolar
changes had a moderate inverse correlation with glycogen-type vacuolar changes (R
2 = -0.68, p = 0.0005). Male Red Snapper collected in the NW region ( χ
2 (1, n = 55)
= 9.53, p = 0.002) and Gulf-wide ( χ
2 (1, n = 150)
=

3.81, p = 0.037) had a higher prevalence of glycogen-type vacuolar change compared to females. Atrophy was more prevalent in females compared to males collected Gulf-wide ( χ
2 (1, n = 150)
= 4.99, p = 0.027)
and in the NC region ( χ
2 (1, n = 91)
= 7.546, p = 0.008). The sex difference
in atrophy was largely driven by the small sample size of males ( n = 2) compared to females ( n = 8). In the NC region, males also had a higher prevalence of bile duct hyperplasia ( χ
2 (1, n = 91)
= 3.86, p = 0.0486) and
myxozoan plasmodia ( χ
2 (1, n = 91)
= 5.490, p = 0.017) compared to
females. No other sex-related differences were detected in the observed pathological changes either within a region or Gulf-wide. There were no correlations between standard length and weight (proxy for age) of Red Snapper with any of the observed MHCs, therefore, precluding age as a significant variable in propelling these changes.

3.8. PAH associations with MHCs
There were no relationships between biliary FACs and pathological changes observed for all data combined. However, there were significant differences within a region. Biliary FACs were moderately to strongly correlated with the prevalence of parasites (R
2 = 0.67, p = 0.0498) in
the NW region and intracellular hemosiderin, lipofuscin and ceroid laden macrophages (R
2 = 0.72, p = 0.008) in the NC region. Addition -
ally, the hepatic PAH concentrations were strongly associated with the prevalence of myxozoan plasmodia in Red Snapper Gulf-wide (R 2 =

0.72, p = 0.001), as well as in the NW (R
2 = 0.76, p = 0.018) and NC (R
2 = 0.64, p = 0.025) regions.

3.9. Regional patterns in MHC prevalence
The NC GoM Red Snapper had a higher prevalence of edema ( χ 2 (2, n =
152)
= 25.33, p < 0.0001) and lipid-type vacuolar change ( χ 2 (2, n = 152)
=

6.79, p = 0.033) compared to those in the NW and SW regions. The most severe cases of lipid- and glycogen-type changes were fish collected in the NC and NW region of the Gulf. Cholangiofibrosis was more prevalent in fish located in the NW ( χ
2 (2, n = 152)
= 6.923, p = 0.031) GoM
compared to those from the NC region.

3.10. MHC patterns in the north central time series
Both decreasing and increasing trends were observed in the various MHCs observed in Red Snapper ( Fig. 5 ). Cholangiofibrosis (-29 %, p =

0.011) and edema (-81 %, p < 0.0001) decreased over time (2011 2017) in fish from the NC region of the GoM (Figure 56a).

Conversely, considerable increases over time were observed for PMAs (60 %, p = 0.008), hepatocellular atrophy ( p < 0.0001), and hemosid - erin, lipofuscin and ceroid laden macrophages (140 %, p = 0.001; Fig. 5 b). No other temporal trends were detected in MHCs in Red Snapper collected in the NC region of the GoM.

4. Discussion
In the GoM, this is the most comprehensive assessment of hep - atobiliary PAH concentrations and microscopic hepatic changes in Red Snapper collected between 2011 and 2017. There were no trends Fig. 5. Microscopic hepatic changes (MHCs) in Red Snapper collected in the north central region of the Gulf of Mexico that are demonstrating decreasing (A) and increasing (B) trends in prevalence over time (2011-2017). The MHC trends with an asterisk are significant.

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
8 observed in the PAH concentrations in hepatic tissues whereas biliary FAC levels declined after 2011, remaining relatively stable until 2015 which was followed by a significant increase in 2017. Additionally, 99 % of the Red Snapper evaluated had, on average, five coinciding MHCs and ranged between one to nine for an individual fish. The most significant changes observed through microscopic evaluation of the Red Snapper hepatic tissues included (in order of decreasing prevalence) lymphocytic inflammation, parasite burdens, PMAs, cholangiofibrosis and hepatic vacuolar changes. While some MHCs decreased over time in Red Snap - per from the NC region, significant increasing trends in PMAs, hemo - siderin, lipofuscin and ceroid laden macrophages and hepatocellular atrophy were observed. Hepatobiliary PAHs were associated with PMAs, parasitic infections and myxozoan plasmodia prevalence.

Considerable increases in mean biliary FACs were observed over time in Red Snapper from both the WFS (2013 to 2017: + 625 %) and the NC region (2015 to 2017: + 162 %). Patterns of increasing PAH expo - sures and concentrations have been observed in various matrices in the NC region following environmental disturbances ( Diercks et al., 2018 ; Perez-Umphrey et al., 2018 ; Pulster et al., 2020a , d ; Snyder, 2020 ; Struch et al., 2019 ; Turner et al., 2019 ). It is likely that the chronic in - puts of petroleum hydrocarbons into the GoM are from many sources including riverine discharge, natural seeps, the resuspension of contaminated sediments (e.g. DWH), leaking infrastructure (e.g. Taylor MC20), and the thousands of accidental spills that occur annually in this region from active infrastructure, extraction and transportation activ - ities ( National Research Council, 2003 ; Kolian et al., 2015 ; Mason et al., 2019 ; Sun et al., 2018 ).

Distinct differences in the compositional profile of PAHs in Red Snapper liver tissues were also noted between 2011 and later years.

Factor analysis revealed the distinction was mainly between LMW and alkylated PAHs (factor 1) and HWM compounds (factor 2), with the HMW PAHs more frequently detected in the 2011 liver samples compared to other years. The HMW PAHs are formed by the incomplete combustion of various fuels and organic matter which are mostly asso - ciated with anthropogenic sources. Thus, the in situ burning of surface oil following DWH is a plausible source for the HMW PAHs detected in the 2011 Red Snapper livers from the NC region. Alkylated PAHs and the LMW compounds were significant in the profiles for all other years, suggesting weathering, metabolic processes and increased bioavail - ability of LMW PAHs. This profile pattern dominated by LMW com - pounds in fish is consistent with other regional and global studies ( El-Kady et al., 2018 ; Murawski et al., 2014 ; Pulster et al., 2020a , b ; Snyder, 2020 ; Struch et al., 2019 ; Zhang et al., 2015 ). In the NC region, male Red Snapper had significantly higher hepatobiliary PAH concen - trations compared to females. Higher contaminant concentrations in male fish has been previously attributed to increased biochemical re - sponses during metabolism (e.g, mixed oxygenase function activity) and higher rates of swimming activity and energy expenditure ( Gray et al., 1991 ; Madenjian et al., 2016 ; Stegeman et al., 1982 ). An additional explanation may be the documented maternal offloading of PAHs by spawning female fishes and has been suggested to occur in Red Snapper located in the NW GoM ( Nicholson, 2019 ).

The weak yet significant associations found between hepatobiliary PAHs and biometrics, including some somatic indices, in this study suggests these variables have a negligible influence on hepatobiliary PAH concentrations. These results also suggest that PAH concentrations are likely not influencing the condition of Red Snapper, which is consistent with the conclusions reached for other species in the GoM ( Pulster et al., 2020a , b ; Snyder, 2020 ). Similar relationships were also observed between PAH metabolites and the condition factor of other species in areas with poor water quality, high levels of eutrophication, sedimentation and elevated PAH concentrations ( Freire et al., 2020 ; Vives et al., 2004 ). Condition factors are commonly used as quantitative indicators of individual fish health and can be influenced by age, sex, season, stage of maturation, feeding status, type and quality of food consumed, amount of fat reserves, muscular development and the overall water quality, and environmental conditions. Although Red Snapper in the NC region were larger in length and weight than those collected in other regions, their condition factors were ≤ 39 % lower than most of the other regions. The declining condition factors, bio - metrics and somatic indices in Red Snapper from the NC region and along the WFS of the GoM may have been the result of annual variability in normal physiological parameters however they could also be a sub-lethal response to an acute stress (e.g., oil spill) or exposure to multiple stressors often experienced in this region (e.g., co-occurring xenobiotics, anoxic conditions, diet shifts, salinity and temperature changes). Declines in growth rates were observed in Red Snapper the first three years post-DWH ( Herdter et al., 2017 ) and the condition factors of Golden Tilefish were still in decline as of 2017 ( Snyder, 2020 ).

These studies add supportive evidence that fishes in the NC region may be expressing sub-lethal responses to environmental stressors. Reduced growth rates or condition factors, are often interpreted as a depletion of energy reserves and commonly observed during periods of stress and therefore considered reliable indicators of stress in fish ( Bonga, 1997 ; Geode and Barton, 1990 ; Schreck et al., 2016 ).

A single fish in the NC region near the Mississippi River Delta had a focal area of hepatocellular hyperplasia. Ito cell hyperplasia and rare FCAs were also observed in Red Snapper from this region which could potentially be a consequence of toxin exposure, although the prevalence was too low to detect correlations with PAHs levels. Exposure to carci - nogenic or estrogenic compounds have been linked to altered hepatic foci, such as FCAs, and are considered to represent the first morpho - logical stage of the neoplastic process in fish ( Boorman et al., 1997 ). Age and sensitivity to chemical insult as well as species inclination for neoplastic transformation should be considered as factors in tumor development.

Compensatory adaptation could also play a role in the toxicity tolerance of the Red Snapper population in the GoM, an environment subjected to anthropogenic pollution and other human-related distur - bances for at least a century. Convincing evidence is provided by thriving populations of Atlantic Killifish (Fundulus heteroclitus) in highly polluted estuaries along the US Atlantic coast ( Whitehead et al., 2017 ).

Rapidly evolved adaptation through the mutation of genes and the alteration of biotransformation pathways has led to the metabolic resistance, reduced chemical sensitivity and increased chemical toler - ance in these populations. Although the short generation times of Atlantic Killifish (~4 years) allows for evolutionary change to parallel environmental change ( Whitehead et al., 2017 ), compensatory adaption may be applicable to longer lived species like Red Snapper (~60 years) and should be further investigated since evolved adaption for chemical tolerance may be accompanied by physiological fitness costs. Therefore, the paucity of neoplasms observed in Red Snapper may be a conse - quence of adaptation to chronic toxicant exposure, age, concentration, frequency or duration of exposure.

More than half of the Red Snapper evaluated had PMAs (57 %), which can serve as an indicator of chronic stress, underlying disease, and/or the presence of parasites. The association between PMAs and stress or tissue damage could be confounded with infection ( Steinel and Bolnick, 2017 ). The ubiquitously distribution of PMAs in vertebrates ensures they provide the first line of cell-mediated host defense against pathogens and can be functionally distinct depending on their precur - sory activation ( Grayfer et al., 2018 ). Furthermore, PMAs are robust factories of cytokines, chemokines and lipid mediators, which act to potentiate and fine-tune inflammatory and adaptive immune responses ( Hodgkinson et al., 2015 ). PMAs play a significant role in the activation of the immune response, antigen processing and immunological memory ( Agius and Roberts, 2003 ). The key functions of PMAs in fish have been classified as (1) immune (humoral and inflammatory responses); (2) metabolic dumps for the storage and destruction of exogenous and endogenous substances; and (3) iron recycling ( Steinel and Bolnick, 2017 ; Wolf and Wolfe, 2005 ; Wolke, 1992 ). Environmental and physi - ological changes can stimulate PMA response, underscoring the E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
9 difficulty in interpreting PMA changes. Nonetheless, there are numerous reports demonstrating the increases in size and frequency of PMAs with pollutants and environmentally stressful conditions, indicating their utility as water quality indicators ( Agius and Roberts, 2003 ).

In the present study, there was an increased prevalence of the he - mosiderin, lipofuscin and ceroid laden macrophages in the hepatic tis - sues of Red Snapper in the NC GoM region which were associated with biliary FAC concentrations. Lipofuscin and/or ceroid accumulation has been observed in fish displaying a wide variety of pathological condi - tions, including nutritional deficiencies, bacterial and viral disease, and toxins ( Agius and Roberts, 2003 ). Studies evaluating superoxide radicals produced by peritoneal macrophages from English sole ( Pleuronectes vetulus ) exposed to polycyclic aromatic compounds may provide support to the presence of oxidative stress and increase in ceroid/lipofuscin pigment in treated animals ( Clemons et al., 1999 ).

Red Snapper in the GoM had a similar frequency of both glycogen- and lipid-type vacuolar change, complicating the interpretation. The loss or accumulation of hepatic glycogen and/or lipid can be a common morphological response to toxicity ( Wolf and Wolfe, 2005 ), although these changes could also be a factor of normal physiological processes (e. g., spawning) or secondary to decreased body condition from starvation, stress, or more typically, concurrent with disease ( Wolf and Wolfe, 2005 ). Both the accumulation or depletion of lipid and glycogen has been observed in fish either as a direct or secondary consequence following exposures to toxins, including PAHs at very low environ - mentally relevant concentrations ( Marty et al., 2003 ; Miranda et al., 2008 ). Hepatic lipid storage in fish is an important energy reserve that can fluctuate seasonally as a response to spawning or periods of star - vation. The high reproductive investment results in the mobilization of hepatic and muscle tissue lipid reserves that are transferred to the go - nads for maturation and spawning. Reproductive assessment and energy status can be accurately represented by the gonadosomatic (GSI) and hepatosomatic (HSI) index, respectively. The majority of the Red Snapper in this study were collected in August, just following their peak spawning season (May-July) in the northern GoM. There was a signifi - cant positive association between the GSI and HSI for Red Snapper collected both Gulf-wide (r = 0.484, p = 0.007) and in the north central region (r = 0.925, p = 0.001), suggesting there does not appear to be a trade-off between energy reserves and gonad maturation and/or spawning. There were, however, significant temporal trends for both the HSI and the GSI for Red Snapper collected in the north central region suggesting changes in energy reserves over time. The elevated glycogen storage could have been replacement energy storage during periods of starvation and/or to compensate for the lipid remobilization during spawning while the elevated hepatic lipids may be the result of rebuilding the energy reserves post-spawning. However, further inves - tigation is needed to determine the true cause of vacuolar change and the declining trends in somatic indices in Red Snapper.

While it is not uncommon to see low levels of parasites in wild caught fish, a pattern of infection was associated with biliary FAC concentra - tions in Red Snapper. The prevalence of myxozoan plasmodia in Red Snapper were also strongly associated with hepatic PAHs in this study.

Nematodes specifically have been frequently described in a number of reef fishes from the GoM, including snappers, groupers and jacks ( Montoya-Mendoza et al., 2017 ; Moravec et al., 2014 ). The prevalence of intermediate hosts, infection level of prey, and host susceptibility play a significant role in parasite infection rates. Fish experiencing decreased physiological condition or exposed to polluted waters tend to suffer from higher levels of parasitic infestations ( Rohlenov ´a et al., 2011 ). A recent study documented the effects of seasonal variability on host immunity and physiology as well as the different parasitic life-strategies that can influence the physiology of fish; activating various immunity pathways ( Rohlenov ´a et al., 2011 ). Some parasites (e.g. Myxidium sp.) are known to induce strong inflammatory responses responsible for tissue deterio - ration and parasite encystation that can interrupt normal organ function and reproductive potential in severe cases ( Al-Jahdali and Hassanine, 2010 ). Additionally, a recent study provides compelling evidence for the strong associations between biliary neoplasia and myxozoan infections in White Perch ( Morone americana ) collected in Chesapeake Bay ( Mat - sche et al., 2020 ).

Inflammation is a physiological process in response to tissue damage caused from infection, chemical irritation, and/or wounding ( Philip et al., 2004 ). Additionally, chronic inflammation that is not caused by infection may also contribute to carcinogenesis ( Philip et al., 2004 ). In Red Snapper, lymphocytic inflammation was observed in 58 % of the samples (Gulf-wide) and was commonly associated with the occurrence of parasitic infections. While both may be common, there is a vast amount of literature demonstrating the direct relationship between chronic inflammation and carcinogenesis ( Lu et al., 2006 ; Philip et al., 2004 ).

In comparison to the NC benthic Golden Tilefish evaluated in a concurrent study ( Snyder, 2020 ), Red Snapper had a lower prevalence of glycogen-type vacuolar change, hepatocellular atrophy and FCAs.

Conversely, Red Snapper had a higher prevalence of lymphocytic inflammation, parasites, PMAs, granulomas, cholangiofibrosis, lipid-type vacuolar change, bile duct hyperplasia, hepatocellular ne - crosis, and Ito cell hyperplasia. These data support species-specific susceptibility to hepatic changes. A parallel study evaluating oxidative stress and immune system biomarkers in both Golden Tilefish and Red Snapper may provide some insight ( Deak, 2020 ). Golden Tilefish from the NC region exhibited differential oxidative stress and immune system biomarker patterns as well as a possible indication of compensatory metabolism. These results combined with the chronic exposures to PAH contaminated sediments, suggest Tilefish may have increased resilience to environmental stressors and thus decreased susceptibility to disease and MHCs.

The biomarker patterns were more variable in Red Snapper with no indication that this species may be less resilient ( Deak, 2020 ), however, there was an increased prevalence of several MHCs in Red Snapper compared to Golden Tilefish. It is possible the chronic exposure to PAHs in the NC region is causing a reduction in the oxidative stress response in Red Snapper through lower expression of the evaluated biomarkers. The reduced oxidative response combined with the increased and persistent PAH exposures in this region will result in cellular damage and a decreased metabolic capacity to react to toxins, which further increases the disease susceptibility in organisms ( Adams and Sonne, 2013 ; Lush - chak, 2011 ; Quintanilla-Mena et al., 2020 ). Additionally, hypoxia-induced DNA methylation and hormone regulation has been observed in Red Snapper ( Rahman and Thomas, 2014 ). High levels of DNA hypomethylation have also been associated with chronic exposures to environmental contaminants, including hydrocarbons, and subse - quently resulting in further genomic changes that have been linked with the development of disease and cancer ( Mirbahai et al., 2011 ; Quinta - nilla-Mena et al., 2020 ).

The wide variety of MHCs observed in the liver tissues of Red Snapper have been described in other fishes, located in the GoM and globally, that were exposed to inorganic and organic pollutants, including PCBs, OCPs, PAHs and metals ( Brown-Peterson et al., 2015 ; Freire et al., 2020 ; Gaber, 2013 ; Miranda et al., 2008 ). However, it is important to note that many of the microscopic changes have also been identified in other exposure and field studies as a result of age, sex, di - etary status, or genetic predisposition. The observed prevalence of MHCs in Red Snapper could be a consequence of normal physiological changes exacerbated by chemical exposures or they could be secondary re - sponses to other environmental stressors. Some of the observed MHCs were associated with hepatobiliary PAH concentrations but require further work to determine the role of PAHs and the induction of these MHCs in this species.

5. Conclusions
This study documents declining biometrics, somatic indices and E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
10 condition factors combined with increasing levels of both PAH exposure and microscopic hepatic changes in Red Snapper from the NC GoM.

Furthermore, 99 % of the Red Snapper evaluated Gulf-wide had one to nine microscopic hepatic changes with an average of five coinciding in an individual fish. While FAC exposures increased significantly in 2017, PAH liver burdens have remained relatively stable, indicating Red Snapper have been able to maintain metabolic clearance thus far.

However, the continued increase in PAH exposures and the associated metabolism required, may have an energetic cost that is manifesting as one or more of the 15 MHCs observed in Red Snapper. The MHCs observed in Red Snapper can be characterized as inflammatory re - sponses, metabolic responses, degenerative responses, nonneoplastic proliferative and neoplastic lesions that could be signaling potential disease progression. Many of the observed responses (i.e., declining growth rates and condition factors) and MHCs in Red Snapper are known secondary responses to stress in fish ( Eissa and Wang, 2016 ).

Considering there is a limited number of ways in which stressors can affect fish populations, the response of adult fish populations to pollu - tion may be indistinguishable from the response to other stressors such as overfishing or eutrophication ( Ryder and Edwards, 1985 ). An over - abundance of co-occurring chemical substances in the aquatic environ - ment, instead an overabundance of co-occurring chemical substances may have additive, synergistic, antagonistic or even potentiated effects.

Pollution can elicit a broad range of physiological responses in the form of lethal, sub-lethal, metabolic or behavioral changes in fish. Subtle endogenous effects and health indices (like those described herein) measure the health of the ecosystem without necessarily being attrib - uted to a particular chemical or specific concentration ( Ryder and Edwards, 1985 ). Hepatic histopathology has been used as a valuable monitoring tool for decades with high ecological relevance, and has been proposed as the most reliable indicator of health impairment among aquatic species ( Hinton et al., 2001 ; Yancheva et al., 2016 ). If we continue to ignore the secondary stress effects and responses observed in fish, then we will never fully understand the impacts and toxicity of chemicals ( Bonga, 1997 ). Consequentially, the abovementioned range of non-neoplastic and neoplastic hepatic toxicities and declining health indices in Red Snapper serve as significant early warning signs and allow for regional comparisons of water quality, environmental health, and the detection of ecosystem trends, all of which can have population level implications.

Author contribution statement
EP assisted in sample acquisition, completed the comprehensive data analysis and wrote the manuscript with contribution from all coauthors.

SF conducted all gross and histological evaluations. BC and JM assisted in sample acquisition and processing. SAM conceived and completed the study design and participated in all sample acquisitions and study oversight.

Declaration of Competing Interest
The authors declare no conflict of interest.

Acknowledgements
We thank the captain and crew of the R/V Weatherbird II and Florida Institute of Oceanography (FIO), S. Gilbert for logistical support and the “Mud & Blood ” field teams from the USA and Mexico. We extend our gratitude to the G. Stuntz lab at Harte Research Institute (T. Topping and J. Wetz) and E. Peebles lab (SHELF Project) at the University of South Florida for additional sample collections. We appreciate the support of the Mexican and USA State Departments to allow international sampling efforts. Our appreciation is also extended to L. Brandenburg and M.

Campbell for laboratory assistance. All sampling was conducted in accordance with Protocols IS00000515 and IS00003388 approved by the Institutional Animal Care and Use Committee at the University of South Florida. This research was supported by grants from The Gulf of Mexico Research Initiative, C-IMAGE I (# SA 12-10), C-IMAGE II (# SA 15-16), and C-IMAGE III (# SA 18-16). Data are publicly available through the Gulf of Mexico Research Initiative Information & Data Cooperative (GRIIDC) at https://data.gulfresearchinitiative.org (doi:

10.7266/n7-nmsy-tq94, 10.7266/RX6DR04 N, and 10.7266/N7 × 34W1J).

Appendix A. Supplementary data
Supplementary material related to this article can be found, in the online version, at doi: https://doi.org/10.1016/j.aquatox.2020.105714 .

References
Adams, D.H., Sonne, C., 2013. Mercury and histopathology of the vulnerable goliath grouper, Epinephelus itajara, in U.S. waters: a multi-tissue approach. Environ. Res. 126, 254 – 263 .

Agius, C., Roberts, R.J., 2003. Melano-macrophage centres and their role in fish pathology. J. Fish Dis. 26, 499 – 509 .

Al-Jahdali, M.O., Hassanine, R., 2010. Ovarian abnormality in a pathological case caused by Myxidium sp (Myxozoa, Myxosporea) in onespot snapper fish Lutjanus monostigma (Teleostei, Lutjanidae) from the Red Sea. Acta Parasitolog. 55, 1 – 7 .

Bonga, S.E.W., 1997. The stress response in fish. Physiol. Rev. 77, 591 – 625 .

Boorman, G.A., Botts, S., Bunton, T.E., Fournie, J.W., Harshbarger, J.C., Hawkins, W.E., Hinton, D.E., Jokinen, M.P., Okihiro, M.S., Wolfe, M.J., 1997. Diagnostic criteria for degenerative, inflammatory, proliferative nonneoplastic and neoplastic liver lesions in medaka (Oryzias latipes): consensus of a national toxicology program pathology working group. Toxicol. Pathol. 25, 202 – 210 .

Brown-Peterson, N.J., Krasnec, M., Takeshita, R., Ryan, C.N., Griffitt, K.J., Lay, C., Mayer, G.D., Bayha, K.M., Hawkins, W.E., Lipton, I., Morris, J., Griffitt, R.J., 2015.

A multiple endpoint analysis of the effects of chronic exposure to sediment contaminated with Deepwater Horizon oil on juvenile Southern flounder and their associated microbiomes. Aquat. Toxicol. 165, 197 – 209 .

Clemons, E., Arkoosh, M.R., Casillas, E., 1999. Enhanced superoxide anion production in activated peritoneal macrophages from English sole (Pleuronectes vetulus) exposed to polycyclic aromatic compounds. Mar. Environ. Res. 47, 71 – 87 .

Collier, T.K., Anulacion, B.F., Arkoosh, M.R., Dietrich, J.P., Incardona, J.P., Johnson, L.

L., Ylitalo, G.M., Myers, M.S., 2013. Effects on Fish of polycyclic aromatic hydrocarbons (PAHS) and naphthenic acid exposures. In: Keith, B., Tierney, A.P.F., Colin, J.B. (Eds.), Fish Physiology: Organic Chemical Toxicology of Fishes. Academic Press, pp. 195 – 255 .

Curtis, L.R., Garzon, C.B., Arkoosh, M., Collier, T., Myers, M.S., Buzitis, J., Hahn, M.E., 2011. Reduced cytochrome P4501A activity and recovery from oxidative stress during subchronic benzo[a]pyrene and benzo[e]pyrene treatment of rainbow trout.

Toxicol. Appl. Pharmacol. 254, 1 – 7 .

Deak, K., 2020. A Baseline Health Evaluation of Golden Tilefish (Lopholatilus Chamaeleonticeps) and Red Snapper (Lutjanus Campechanus), Post-deepwater Horizon College of Marine Science. University of South Florida, Tampa, FL .

Diercks, A.R., Dike, C., Asper, V.L., DiMarco, S.F., Chanton, J.P., Passow, U., 2018. Scales of seafloor sediment resuspension in the northern Gulf of Mexico. Elementa-Sci.

Anthropocene 6 .

Eissa, N., Wang, H.-P., 2016. Transcriptional stress responses to environmental and husbandry stressors in aquaculture species. Rev. Aquac. 8, 61 – 88 .

El-Kady, A.A., Wade, T.L., Sweet, S.T., 2018. Assessment and ecological indicators of total and polycyclic aromatic hydrocarbons in the aquatic environment of lake Manzala, Egypt. J. Environ. Sci. Health, Part A 53, 854 – 865 .

Freire, M.M., Amorim, L.M.F., Buch, A.C., Gonçalves, A.D., Sella, S.M., Cassella, R.J., Moreira, J.C., Silva-Filho, E.V., 2020. Polycyclic aromatic hydrocarbons in bays of the Rio de Janeiro state coast, SE - Brazil: effects on catfishes. Environ. Res. 181, 108959 .

Gaber, H.S., 2013. Fish health as a biomarker for the condition of Lake Nasser (Review article). J. Appl. Sci. Res. 9, 5794 – 5810 .

Geode, R.W., Barton, B.A., 1990. Organismic indices and an autopsy-based assessment as indicators of health and condition of fish. In: Adams, S.M. (Ed.), Biological Indicators of Stress in Fish. American Fisheries Symposium. Bethesda, MD .

Gingerich, W.H., 1982. Hepatic toxicology of fishes. In: Weber, L. (Ed.), Aquatic Toxicology. Raven Press, New York, pp. 55 – 105 .

Gray, E.S., Woodin, B.R., Stegeman, J.J., 1991. Sex differences in hepatic monooxygenases in winter flounder (Pseudopleuronectes americanus) and scup (Stenotomus chrysops) and regulation of P450 forms by estradiol. J. Exp. Zool. 259, 330 – 342 .

Grayfer, L., Kerimoglu, B., Yaparla, A., Hodgkinson, J.W., Xie, J., Belosevic, M., 2018.

Mechanisms of fish macrophage antimicrobial immunity. Front. Immunol. 9, 1105- 1105 .

Herdter, E.S., Chambers, D.P., Stallings, C.D., Murawski, S.A., 2017. Did the Deepwater Horizon oil spill affect growth of Red Snapper in the Gulf of Mexico? Fish. Res. 191, 60 – 68 .

E.L. Pulster et al.

Aquatic Toxicology 230 (2021) 105714
11 Hinton, D.E., Segner, H., Braunbeck, T., 2001. Toxic responses of the liver. In:

Schlenk, D., Benson, W.H. (Eds.), Target Organ Toxicity in Marine and Freshwater Teleosts. Taylor & Francis, New York, NY .

Hodgkinson, J.W., Grayfer, L., Belosevic, M., 2015. Biology of bony fish macrophages.

Biology (Basel) 4, 881 – 906 .

Katsumiti, A., Domingos, F.X.V., Azevedo, M., da Silva, M.D., Damian, R.C., Almeida, M.

I.M., de Assis, H.C.S., Cestari, M.M., Randi, M.A.F., Ribeiro, C.A.O., Freire, C.A., 2008. An assessment of acute biomarker responses in the demersal catfish Cathorops spixii after the Vicu ˜na Oil Spill in a harbour estuarine area in Southern Brazil.

Environ. Monit. Assess. 152, 209 .

Kolian, S.R., Porter, S.A., Sammarco, P.W., Birkholz, D., Cake, E.W., Subra, W.A., 2015.

Oil in the Gulf of Mexico after the capping of the BP/Deepwater Horizon Mississippi Canyon (MC-252) well. Environ. Sci. Pollut. Res. 22, 12073 – 12082 .

Lu, H., Ouyang, W., Huang, C., 2006. Inflammation, a key event in Cancer development.

Mol. Cancer Res. 4, 221 – 233 .

Lushchak, V.I., 2011. Environmentally induced oxidative stress in aquatic animals.

Aquat. Toxicol. 101, 13 – 30 .

Madenjian, C.P., Rediske, R.R., Krabbenhoft, D.P., Stapanian, M.A., Chernyak, S.M., O ’ Keefe, J.P., 2016. Sex differences in contaminant concentrations of fish: a synthesis. Biol. Sex Differ. 7 .

Marty, G.D., Hoffmann, A., Okihiro, M.S., Hepler, K., Hanes, D., 2003. Retrospective analysis: bile hydrocarbons and histopathology of demersal rockfish in Prince William Sound, Alaska, after the Exxon Valdez oil spill. Mar. Environ. Res. 56, 569 – 584 .

Mason, A.L., Taylor, J.C., MacDonald, I.R., 2019. An Integrated Assessment of Oil and Gas Release Into the Marine Environment at the Former Taylor Energy MC20 Site.

NOAA National Ocean Service, National Centers for Coastal Ocean Science. Silver Spring, MD, p. 147 .

Matsche, M.A., Blazer, V.S., Pulster, E.L., Mazik, P.M., 2020. High prevalence of biliary neoplasia in white perch Morone americana from Chesapeake Bay, Maryland, USA:

Potential roles of bile duct parasites and environmental contaminants. Dis. Aquat.

Org.

Miranda, A.L., Roche, H., Randi, M.A.F., Menezes, M.L., Ribeiro, C.A.O., 2008.

Bioaccumulation of chlorinated pesticides and PCBs in the tropical freshwater fish Hoplias malabaricus: histopathological, physiological, and immunological findings.

Environ. Int. 34, 939 – 949 .

Mirbahai, L., Yin, G., Bignell, J.P., Li, N., Williams, T.D., Chipman, J.K., 2011. DNA methylation in liver tumorigenesis in fish from the environment. Epigenetics 6, 1319 – 1333 .

Montoya-Mendoza, J., Salgado-Maldonado, G., Favila-Castillo, M.E., Vazquez- Hurtado, G., del Refugio Castaneda-Chavez, M., 2017. Communitities of helminth parasites in five Carangidae species from the coast of Veracruz, Mexico, southern Gulf of Mexico. Glob. J. Sci. Front. Res. C Biol. Sci. 17 .

Moravec, F., Bakenhaster, M., Fajer-Avila, E.J., 2014. Three new gonad-infecting species of Philometra (Nematoda: Philometridae) parasitic in Lutjanus spp. (Lutjanidae) in the northern Gulf of Mexico off Florida, USA. Folia Parasitol. 61, 355 – 369 .

Murawski, S.A., Hogarth, W.T., Peebles, E.B., Barbeiri, L., 2014. Prevalence of external skin lesions and polycyclic aromatic hydrocarbon concentrations in Gulf of Mexico fishes, post-deepwater horizon. Trans. Am. Fish. Soc. 143, 1084 – 1097 .

Murawski, S.A., Peebles, E.B., Gracia, A., Tunnell, J.W., Armenteros, M., 2018.

Comparative abundance, species composition, and demographics of continental shelf fish assemblages throughout the Gulf of Mexico. Mar. Coast. Fish. Dyn. Manag.

Ecosyst. Sci. 10, 325 – 346 .

National Research Council, 2003. Oil in the Sea III:Inputs, Fates, and Effects. The National Academies Press, Washington, D.C, ISBN 0-309-08438-5 .

Nicholson, T.J., 2019. PAH Exposure in Red Snapper (Lutjanus Campechanus) Collected at Natural and Artifical Reef Systems in the Northwestern Gulf of Mexico, department of Biological Sciences. Universtiy of South Florida, St. Petersburg, FL .

Oliva, M., Vicente-Martorell, J.J., Galindo-Ria ˜no, M.D., Perales, J.A., 2013.

Histopathological alterations in Senegal sole, Solea senegalensis, from a polluted Huelva estuary (SW, Spain). Fish Physiol. Biochem. 39, 523 – 545 .

Patterson, W.F., Cowan, J.H., Wilson, C.A., Shipp, R.L., 2001. Age and growth of red snapper, Lutjanus campechanus, from an artificial reef area off Alabama in the northern Gulf of Mexico. Fish. Bull. (Wash. D. C.) 99, 617 – 627 .

Perez-Umphrey, A.A., Bergeon Burns, C.M., Stouffer, P.C., Woltmann, S., Taylor, S.S., 2018. Polycyclic aromatic hydrocarbon exposure in seaside sparrows ( Ammodramus maritimus ) following the 2010 Deepwater Horizon oil spill. Sci. Total Environ. 630, 1086 – 1094 .

Philip, M., Rowley, D.A., Schreiber, H., 2004. Inflammation as a tumor promoter in cancer induction. Semin. Cancer Biol. 14, 433 – 439 .

Pulster, E.L., Gracia, A., Armenteros, M., Carr, B.E., Mrowicki, J., Murawski, S.A., 2020a.

Chronic PAH exposures and associated declines in fish health indices observed for ten grouper species in the Gulf of Mexico. Sci. Total Environ. 703, 135551 .

Pulster, E.L., Gracia, A., Armenteros, M., Toro-Farmer, G., Snyder, S.M., Carr, B.E., Schwaab, M.R., Nicholson, T.J., Mrowicki, J., Murawski, S.A., 2020b. A first comprehensive baseline of hydrocarbon pollution in Gulf of Mexico fishes. Nat. Sci.

Reports .

Pulster, E.L., Gracia, A., Snyder, S.M., Deak, K., Fogelson, S.B., Murawski, S.A., 2020c.

Chronic sub-lethal effects observed in wild-caught fishes following two major oil spills in the Gulf of Mexico: Deepwater Horizon and Ixoc 1. In: Murawski, S.A., Ainsworth, C.H., Gilbert, S., Hollander, D.J., Paris, C.B., Schluter, M., Wetzel, D.L. (Eds.), Deep Oil Spills: Facts, Fate and Effects. Springer Nature, Switzerland, AG .

Pulster, E.L., Gracia, A., Snyder, S.M., Romero, I.C., Carr, B., Toro-Farmer, G.A., Murawski, S.A., 2020d. Polycyclic aromatic hydrocarbon baselines in Gulf of Mexico fishes. In: Murawski, S.A., Ainsworth, C.H., Gilbert, S., Hollander, D.J., Paris, C.B., Shluter, M., Wetzel, D.L. (Eds.), Scenarios and Responses to Future Deep Oil Spills:

Fighting the Next War. Springer Nature, Switzerland AG .

Quintanilla-Mena, M., Gold-Bouchot, G., Zapata-P ´erez, O., Rubio-Pi ˜na, J., Quiroz- Moreno, A., Vidal-Martínez, V.M., Aguirre-Macedo, M.L., Puch-Hau, C., 2020.

Biological responses of shoal flounder (Syacium gunteri) to toxic environmental pollutants from the southern Gulf of Mexico. Environ. Pollut. 258, 113669 .

Rahman, M.S., Thomas, P., 2014. Hypoxia-induced global DNA methylation and IGF regulation in a marine teleost, red snapper. Integr. Comp. Biol. 54. E335-E335 .

Reddy, A.P., Spitsbergen, J.M., Mathews, C., Hendricks, J.D., Bailey, G.S., 1999.

Experimental hepatic tumorgenicity by environmental hydrocarbon dibenzo[ a,l ] pyrene. J. Environ. Pathol. Toxicol. Oncol. 18, 261 – 269 .

Rohlenov ´a, K., Morand, S., Hyr ˇsl, P., Tolarov ´a, S., Flaj ˇshans, M., ˇ
Simkov ´a, A., 2011. Are
fish immune systems really affected by parasites? An immunoecological study of common carp (Cyprinus carpio). Parasit. Vectors 4, 120 .

Ryder, R.A., Edwards, C.J., 1985. Conceptual Approach for the Application of Biological Indicators of Ecosystem Quality in the Great Lakes Basin. International Joint Commission, Great Lakes Regional Office, Ontario. Ministry of Natural Resources.

Fisheries Branch, Great Lakes Fishery Commission .

Schreck, C.B., Tort, L., Farrell, A.P., Brauner, C.J., 2016. Biology of Stress in Fish.

Academic Press, Cambridge, MA, USA .

Snyder, S.M., 2020. Polycyclic Aromatic Hydrocarbon Exposure, Hepatic Accumulation, and Associated Microscopic Hepatic Changes in Gulf of Mexico Tilefish (Lopholatilus chamaeleonticeps), College of Marine Science. University of South Florida, Tampa, FL, p. 191 .

Snyder, S.M., Pulster, E.L., Wetzel, D.L., Murawski, S.A., 2015. PAH exposure in Gulf of Mexico demersal fishes, post-deepwater horizon. Environ. Sci. Technol. 49, 8786 – 8795 .

Stegeman, J.J., Pajor, A.M., Thomas, P., 1982. Influence of estradiol and testosterone on cytochrome P-450 and monooxygenase activity in immature brook trout, Salvelinus fontinalis. Biochem. Pharmacol. 31, 3979 – 3989 .

Steinel, N.C., Bolnick, D.I., 2017. Melanomacrophage centers As a histological indicator of immune function in fish and other poikilotherms. Front. Immunol. 8 .

Struch, R., Pulster, E.L., Schreier, A.M., Murawski, S.A., 2019. Hepatobiliary analyses suggest chronic PAH exposure in Hakes (Urophycis spp.) following the deepwater horizon oil spill. Environ. Toxicol. Chem. In Press .

Sun, S., Hu, C., Garcia-Pineda, O., Kourafalou, V., Le H ´enaff, M., Androulidakis, Y., 2018.

Remote sensing assessment of oil spills near a damaged platform in the Gulf of Mexico. Mar. Pollut. Bull. 136, 141 – 151 .

Tarnecki, J.H., Patterson, W.F., 2015. Changes in red snapper diet and trophic ecology following the deepwater horizon oil spill. Mar. Coast. Fish. Dyn. Manag. Ecosyst. Sci. 7, 135 – 147 .

Turner, R.E., Rabalais, N.N., Overton, E.B., Meyer, B.M., McClenachan, G., Swenson, E.

M., Besonen, M., Parsons, M.L., Zingre, J., 2019. Oiling of the continental shelf and coastal marshes over eight years after the 2010 Deepwater Horizon oil spill. Environ.

Pollut. 252, 1367 – 1376 .

Varanasi, U., Stein, J.E., Nishimoto, M., Reichert, W.L., Collier, T.K., 1987. Chemical carcinogenesis in feral fish: uptake, activation, and detoxification of organic xenobiotics. Environ. Health Perspect. 71, 155 – 170 .

Vives, I., Grimalt, J.O., Fern ´andez, P., Rosseland, B., 2004. Polycyclic aromatic hydrocarbons in fish from remote and high mountain lakes in Europe and Greenland.

Sci. Total Environ. 324, 67 – 77 .

Whitehead, A., Dubansky, B., Bodinier, C., Garcia, T.I., Miles, S., Pilley, C., Raghunathan, V., Roach, J.L., Walker, N., Walter, R.B., Rice, C.D., Galvez, F., 2012.

Genomic and physiological footprint of the Deepwater Horizon oil spill on resident marsh fishes. Proc. Natl. Acad. Sci. U.S.A. 109, 20298 – 20302 .

Whitehead, A., Clark, B.W., Reid, N.M., Hahn, M.E., Nacci, D., 2017. When evolution is the solution to pollution: key principles, and lessons from rapid repeated adaptation of killifish (Fundulus heteroclitus) populations. Evol. Appl. 10, 762 – 783 .

Wolf, J.C., Wheeler, J.R., 2018. A critical review of histopathological findings associated with endocrine and non-endocrine hepatic toxicity in fish models. Aquat. Toxicol. 197, 60 – 78 .

Wolf, J.C., Wolfe, M.J., 2005. A brief overview of nonneoplastic hepatic toxicity in fish.

Toxicol. Pathol. 33, 75 – 85 .

Wolke, R.E., 1992. Piscine macrophage aggregates: a review. Annu. Rev. Fish Dis. 2, 91 – 108 .

Yancheva, V., Velcheva, I., Stoyanova, S., Georgieva, E., 2016. Histological biomarkers in fish as a tool in ecological risk assessment and monitoring programs: a review. Appl.

Ecol. Environ. Res. 14, 47 – 75 .

Zhang, X., Xia, X., Li, H., Zhu, B., Dong, J., 2015. Bioavailability of pyrene associated with suspended sediment of different grain sizes to Daphnia magna as investigated by passive dosing devices. Environ. Sci. Technol. 49, 10127 – 10135 .

E.L. Pulster et al.