This decision sets out the Court's findings of fact and conclusions of law pursuant to Federal Rule of Civil Procedure 52 following a three-day bench trial. Each party is a maritime insurer. The central issue at trial concerned responsibility for costs arising out of the grounding and ensuing salvage of two oil barges in the Mississippi River in April 2014.
Plaintiff is Starr Indemnity & Liability Company ("Starr"). Starr's insured, Genesis Marine LLC ("Genesis"), incurred salvage costs after two oil barges it owned and operated were grounded in the Mississippi. Starr reimbursed Genesis for those costs. Now, suing in admiralty on behalf of Genesis, which assigned Starr its right to so sue, Starr brings a single claim of breach of contract against another of Genesis's insurers, defendant Water Quality Insurance Syndicate ("WQIS"), which declined to reimburse Genesis for its salvage costs. Starr claims that WQIS was responsible for covering these costs under several provisions of a policy under which WQIS had insured Genesis for costs incurred as the result of oil spills and incidents giving rise to a "substantial threat" of an oil spill. Although the grounding of Genesis's barges undisputedly did not cause any spill, Starr claims that the barges' grounding presented a substantial threat of a *553discharge, so as to trigger the WQIS policy. WQIS disputes this.
Trial was held between January 5 and 9, 2018. The Court heard testimony from five live witnesses, each of whose direct testimony was received in the form of a sworn affirmation and who was subject to cross- and redirect examination. Two were called by Starr: Paul Ferguson, a Starr vice president, see PX 23 ("Ferguson Aff."); and J. Kenneth Edgar, an expert witness, see PX 25 ("Edgar Aff."). Three were called by WQIS: Robert Schrader, a manager of response services for Gallagher Marine Systems LLC, see DX 72 ("Schrader Decl."); Larry Diamond, a WQIS vice president (and a 30(b)(6) witness), see DX 73 ("Diamond Decl."), Dkt. 74 ("Diamond Supp. Decl."); and George Randall, an expert witness, see DX 78 ("Randall Decl.").
In addition, the Court received written testimony-pursuant to the U.S. Coast Guard's Tuohy regulations-of U.S. Coast Guard Chief Petty Officer Heather Norman, in the form of a sworn declaration, see PX 22 ("Norman Decl.").1 The Court also received the testimony of Karen Pape, a Starr witness, by affirmation, see PX 21 ("Pape Affirmation"); and designations of deposition testimony of Pape and four others: Deeann Ebanks, the Rule 30(b)(6) deponent for T & T Salvage LLC; Chuck Pennington, the on-site surveyor for Merrill Marine during the incident; Joanie Murphy; and John Moy. The Court also received the parties' Joint Stipulation of Undisputed Facts, Dkt. 64 ("JSF"), the parties' joint exhibits ("JX"), and exhibits offered by each side ("PX" and "DX").2
The findings of fact that follow are based on the Court's review of the entire trial record. Where based in whole or part on a witness's testimony, the Court's findings reflect credibility determinations based on the Court's assessment of, inter alia , the relevant witness or witnesses' experience, knowledge, and demeanor.
For the reasons set forth below, the Court finds that the barges did not pose a substantial threat of discharge; that the salvage efforts at issue were not undertaken for the purpose of mitigating such a discharge; and that, had Genesis not undertaken the salvage efforts it did, the Coast Guard would not have ordered Genesis to do so. The Court therefore holds that none of the provisions of WQIS's insurance policy were triggered, and that this policy, therefore, does not cover the costs Genesis incurred.
I. Findings of Fact
A. The Insurers, the Insured, and the Barges
Starr is a Texas insurance company with its principal place of business in New York. JSF ¶ 1. WQIS is an unincorporated association of marine insurers with its principal place of business in New York. Id. ¶ 2.
*554Genesis is a limited liability company with its principal place of business in Houston, Texas. JSF ¶ 3. Karen Pape was at all relevant times the Senior Vice President and Controller of Genesis's affiliate, Genesis Energy LLC. JSF ¶ 4. Pape was responsible for placing Genesis's insurance policies. Id.
Genesis was and is the owner of the barges at issue in this case: the GM-5001 and the GM-5002. JSF ¶¶ 12-13. Both barges are double-skinned tank barges designed to carry liquid in bulk in six cargo tanks on each barge. See DX 26 at 1; DX 27 at 1; Tr. 287 (Diamond). Each barge had cargo tanks at the center of the barge, with empty "wing tanks" below (along the hull) and on the sides. See DX 26 at 1; DX 27 at 1; Tr. 123-25 (Edgar). The wing tanks protect the cargo in the central cargo tanks and provide buoyancy or ballast for the barges. Tr. 125 (Edgar). Each barge was approximately 297 feet, 6 inches long, with a breadth of 54 feet, and a depth of 12 feet. JX C at 5.
At the time of the grounding, the barges "were in all respects tight, staunch, strong, equipped and supplied and in all respects seaworthy and fit for the service for which they were engaged." JSF ¶ 20. GM-5001 was carrying 16,701.8 barrels and GM-5002 was carrying 14,841.3 barrels of decant oil. JSF ¶ 22. The barges were being pushed in an "assembled tow"-that is, with the two barges joined together and a single tug, the M/V Karen Pape, pushing them downstream. JSF ¶¶ 17, 18. The GM-5002 was the lead barge and was a raked barge, meaning its bow curved up from the waterline. JSF ¶¶ 13, 18. The GM-5001 was the aft barge and was oriented with its bow-although an unraked bow-upstream. JSF ¶ 18. The M/V Karen Pape, a twin-screw towboat also owned by Genesis Marine, was pushing the two-barge tow. JSF ¶ 17.
B. The Insurance Policies
1. Starr's Policies
The M/V Karen Pape and both barges were covered by two Starr policies: a "Hull & Machinery" Policy (MASIHHS00070813) and a "Protection & Indemnity" Policy (MASIHHS00070913). JSF ¶ 7; see JX A (the "Starr Policies"); JSF ¶ 10. Both had effective dates of-i.e. , they covered the period-between July 18, 2013 and July 18, 2014. JSF ¶ 7; Starr Policies ¶ 4. Those policies insured Genesis against losses to its vessels, see Starr Policies at 12-14, 16, and for damage to others' property and personal injury, id. at 22. The Policies exclude coverage for costs imposed on Genesis "directly or indirectly, in consequence of, or with respect to, the actual or potential discharge, emission, spillage, or leakage upon or into the seas, waters, land or air, of oil, petroleum products, chemicals or other substances of any kind or nature whatsoever." Starr Policies at 19-20 ¶ 16; see id. at 26 ¶ 25. That "pollution exclusion" does not apply to losses and costs caused by actual discharges of oil. See Starr Policies at 26-27 ¶ 25.3
2. WQIS's Policy
The M/V Karen Pape and the two barges were also covered by a primary pollution liability insurance policy issued by WQIS, Policy No. 46-80019. JSF ¶¶ 8, 10; see JX B (the "WQIS Policy"). The WQIS Policy also has effective dates of between July 18, 2013 and July 18, 2014. JSF ¶ 8.
This case implicates three provisions of the WQIS Policy.
*555First, section A, subsection (1) of the WQIS Policy insured Genesis for
Liability to the United States or to any Claimant imposed under Subchapter I of [the Oil Pollution Act of 1990 ("OPA") ] and costs and expenses incurred by [Genesis] for Removal and damages under the OPA for which liability would have been imposed under Subchapter I, had [Genesis] not incurred such liability voluntarily.
WQIS Policy at 4. As discussed below, Subchapter I of OPA imposes liability, on parties responsible for vessels, for removal costs and damages resulting from the discharge or substantial threat of discharge of oil. See 33 U.S.C. § 2702(a).
Two provisions within Section A, subsection (7)-"Salvage, Cleaning, Offloading and Miscellaneous Liability"-are also implicated.
Subsection 7(a) insured Genesis for
Costs and expenses incurred by [Genesis] for firefighting, salvage or removal of wreck or debris of any Vessel(s) or cargo carried aboard any such Vessel(s), to the extent that such actions were undertaken for the purpose of stopping a discharge or release, or mitigating or preventing a substantial threat of a discharge under OPA
....
WQIS Policy at 5.
Finally, subsection 7(d) provides
In the event of an Occurrence covered under Section A(1) or (2) of Part I, and the Assured is required to offload Oil or Hazardous Substances carried aboard any Vessel(s) involved in the Occurrence and such requirement is at the order of the United States Coast Guard or designated Federal On-Scene Coordinator, this Policy is extended to indemnify [Genesis] for costs and expenses incurred by [Genesis] for actions taken with the prior consent of WQIS to offload any Oil or Hazardous Substances carried aboard the Vessel(s); and it is also understood and agreed that, in the event of a claim covered hereunder, the Policy will indemnify [Genesis] for (i) costs and expenses incurred by [Genesis] for actions taken with the prior consent of WQIS for the disposal of any contaminated Oil and/or Hazardous Substances offloaded from the Vessel(s); and (ii) additional costs and expenses incurred by [Genesis] for the delivery of offloaded Oil or Hazardous Substances provided, however, that such costs and expenses are incurred with the prior consent of WQIS and would not have been incurred by [Genesis] had the offloading not been required
....
WQIS Policy at 5.
C. The April 6 Grounding of Genesis's Barges
The events giving rise to this suit began on April 6, 2014, when barges GM-5001 and GM-5002 ran aground near Cape Girardeau, Missouri, on the Mississippi River. JSF ¶ 17. Rounding a bend in the river at approximately 7:30 p.m., the captain of the M/V-Pape, pushing the barges downriver, cut the comer too tightly and ran aground on a sandbar outside the marked channel of the river. See PX 1, United States Coast Guard, Report of Investigation into the Circumstances Surrounding the Incident Involving Karen Pape/Grounding on 03/24/2014 at 10, 23 ("Coast Guard Report").4 At the time of the grounding, the barges were aground 9 inches into the bar, *556Coast Guard Report at 23, and the water level in the Mississippi was approximately 25.7 feet, JSF ¶ 24. Because the barges were grounded on a sandbar outside the river's navigable channel, they did not present an obstacle to river traffic. JSF ¶ 21.
As soon as the barges ran aground, the captain of the M/V Karen Pape attempted to dislodge the tow from the sandbar. Coast Guard Report at 23. Despite his several attempts to do so, the barges remained lodged on the bar. Id. During this initial effort to unground the barges, the water level on the river "was falling out at a rate of 4 inches every six hours," thus leaving the barges "harder aground by the minute." Id.
D. The Salvage Operation
On April 6, shortly after the grounding, the captain of the M/V Karen Pape notified the Coast Guard's Marine Safety Unit Paducah ("MSU Paducah"), a division of the Sector Ohio Valley (and itself a unit of the Coast Guard's Eighth District) of the grounding. JSF ¶ 25; see Coast Guard Report at 23. The captain of the M/V Karen Pape continued to attempt to dislodge the barges throughout the following two days, see Coast Guard Report at 46, but without success, see id. at 47.
On April 8, Genesis's insurance broker, McGriff, Seibels & Williams of Missouri, Inc. ("McGriff") notified WQIS of the grounding. JSF ¶ 26; see JSF ¶ 5. That day, Genesis's broker appointed Merrill Marine, an independent marine surveyor, to monitor operations at the site of the grounding. JSF ¶ 27; see Tr. 78 (Ferguson). Merrill Marine's on-site surveyor was Chuck Pennington, who arrived on April 9 and remained on-site intermittently until April 24. JSF ¶ 27.
On April 9 and 10, several other parties arrived on the scene. Under a pre-existing agreement, Genesis had designated T & T Salvage LLC ("T & T") as its salvage responder. See JSF ¶ 29. On April 8, Genesis activated its agreement with T & T Salvage, which, over the next two days, drafted a salvage plan and submitted that plan for approval to the Coast Guard. JSF ¶ 32; see JX C ("April 9 Salvage Plan"). On April 10, the Coast Guard approved the April 9 Salvage Plan. JSF ¶ 32. On April 9, Rob Schrader, working for Maritime Alliance Group, whom WQIS had appointed to represent it on site, arrived on the scene. JSF ¶ 28. He was on site until April 24, when he was replaced by David O'Dougherty, who was on site until May 1. Id.
Throughout its time on site, T & T prepared daily reports, which detailed the salvor's assessment of the barges and summarized the daily activity at the site. See PX 8 ("T & T Daily Reports"). On April 9, T & T's first day on the scene, the daily report indicates that the barges were "in a stable condition resting on [a] muddy bottom." T & T Daily Reports, 4/9/14 at 1. That report noted that no "[s]agging[,] hogging or torque was observed," and that there were no "signs of damage or hull deterioration." Id.5
1. April 9: The Initial Lightering Plan
The April 9 Salvage Plan set out T & T's initial plan for lightering (transferring the oil cargo from) and refloating the barges. See April 9 Salvage Plan at 8. The initial plan noted that, as of T & T's initial assessment of April 9 at 11:30 a.m., no "damage or leakage of oil ha[d] been observed," but that "[o]il boom ha[d] been deployed as *557a precaution." Id. at 3; see id. at 6.6 That initial assessment revealed no "signs of damage or hull deterioration," including no "[s]agging[,] hogging[,] or torque." Id. at 6.
Later that day, at approximately 1:30 p.m., T & T's naval architect arrived on scene. Id. Both T & T's salvage master, Peter Drummond, and its naval architect, Matthew Cooke, Jr., reported observing "no obvious signs that hull integrity ha[d] been compromised." Id. However, the initial assessment identified "minimal structural concerns due to the unsupported cantilevered bow" of one of the barges, which "support[ed] lightering from the bow first." Id. at 7.
T & T developed its initial salvage plan after an "initial damage assessment and discussions with Genesis Marine personnel, the [Coast Guard] and the attending surveyor." Id. at 8. The plan called for lightering the entirety of the oil on board the barges to "reduce the threat of pollution and ensure safe delivery" of the oil. Id. Lightering of the GM-5002 would proceed first, with the GM-5001 to follow. Id. at 9.7 For both barges, lightering would be conducted with the aid of spud (or spar) barges. Id. at 8-9. MSU Paducah reviewed and approved the plan. See Coast Guard Report at 50; JSF ¶ 32.
Spud/spar barges are barges used as work platforms; they have "spuds" or "spars" that descend to the river bottom to support and stabilize the barge. See Tr. 165 (Edgar). The initial lightering plan called for three spud barges to be arranged in a row, perpendicular to the grounded barges, with hoses running across those three spud barges to connect the grounded barges to a receiving barge. April 9 Salvage Plan at 8; Tr. 166 (Edgar). The spud barges would provide a connection between the stranded barges and a receiving barge, which could remain in the main channel of the river. The spud barges and the receiving barge, therefore, would not be at risk of grounding. Tr. 166 (Edgar).
2. April 11: Lightering Proceeds and Then Is Temporarily Stopped
On the morning of April 11, T & T began lightering the GM-5002 barge in just that way. T & T Daily Reports, 4/11/18 at 3; see Tr. 166 (Edgar). T & T's daily report from April 11 reported the barges in the same condition as the previous two days: resting stably on a sandy bottom without visible hogging, sagging, or torque. T & T Daily Reports, 4/11/18 at 1.
T & T began its activities on April 11 with a 7:35 a.m. safety meeting. T & T Daily Reports, 4/11/18 at 3. By 10:20 a.m., the spud barges were in position, and by 10:35 a.m., the hose used to carry oil from the stranded barges to the receiving barges was in place. Id. At 11:12 a.m., the hose was tested "with [the Coast Guard] in attendance." Id. That test was complete by 12:50 p.m. Id. A containment boom was then deployed by "pollution control"-that is, SWS. Id. At 4:30 p.m., pumping began. Id. ; see also Tr. 167 (Edgar). From 4:30 p.m. to 5:40 p.m., pumping continued, with the pressure increasing to 70 psi. T & T Daily Reports, 4/11/18 at 3.
*558At 5:42 p.m., however, T & T began evaluating the "dropping river stage and [the] rate of pumping."Id. At 5:45 p.m., T & T determined that it would continue pumping "so long as clearance below the barges allowed." Id. All the while, T & T began "developing plans" in the event that the "current river stage did not allow continuation of activities." Id. By 6:20 p.m., the water level had fallen enough that the "[m]idle spar barge [was] almost touching bottom," and by 6:33 p.m., the spar barge was "in danger of grounding." Id. at 3-4. Accordingly, T & T began the "procedure to stop pumping" and notified the Coast Guard that it was stopping. Id. at 4.
A factual dispute arose at trial as to whether, as WQIS claims, T & T Salvage stopped lightering the GM-5002 on April 11 because of its concern that the spud barges would become grounded, or whether, as Starr contends, because of concerns about the structural integrity of the GM-5002. Starr claims that lightering was halted because the process of lightering the GM-5002 had begun to cause a section of the barge-the hull outside of an empty ballast tank known as a "wing tank"-to buckle. In Starr's version of events, it was the buckling of this wing tank, the No. 3 Wing Tank-that resulted in the Coast Guard assigning the salvage operation a project number with the National Pollution Funds Center. See Pl. Br. at 10-11. WQIS argued instead that the April 11 lightering was halted because of a concern that the falling water level in the river would leave the deck barges and receiving barge grounded. In WQIS's account, as the water level continued to fall on April 11, there was a risk that the deck barges and receiving barge would also run aground. WQIS argues that it was this risk-that the other barges would ground, too-that prompted T & T Salvage to cease operations on April 11. Def. Br. at 11-12.
The Court finds with WQIS on this point. The Court finds that lightering was not halted because of buckling in the wing tank. (Indeed, it is not clear that Starr, following trial, continues to advance the argument that buckling in the wing tank precipitated the abrupt halting of the lightering on April 11: In its post-trial briefing, Starr no longer purports to explain why the wing tank began to buckle, contending instead merely that it did in fact begin buckling at this time. See Pl. Br. at 41.) In any event, even on Starr's version of events, discussed immediately below, the Coast Guard identified the buckling in the wing tank at 3:30 a.m. on April 12 , some nine hours after T & T Salvage had stopped lightering the GM-5002. Thus, even if Starr's view were credited that the Coast Guard had identified the buckling in the wing tank at 3:30 a.m. on April 12, see infra , it would not follow-and there is no affirmative evidence to suggest-that buckling caused anyone to deviate from the initial lightering plan on April 11 .8
The parties also dispute when the damage to the No. 3 Wing Tank was first discovered: WQIS, relying on T & T's daily report's activities log, argues that T & T's inspection of the barges on April 12, at 9 a.m., done along with the Coast Guard, first revealed the damage. Id. at 2. Starr *559contends, however, that the damage to the wing tank had first been discovered by the Coast Guard at 3:30 a.m. that morning, some five and a half hours earlier. See Pl. Br. at 10-11 & n.6 (citing Coast Guard Report at 51).
Here, too, the Court finds with WQIS. The Court finds it more likely than not that the damage to the wing tank was first discovered during the 9 a.m. inspection and not, as Starr would have it, hours earlier, at 3:30 a.m. The Court makes this finding based on the lack of any corroborating evidence suggesting that the Coast Guard was on board the GM-5002 at 3:30 a.m. The Court also finds it improbable that T & T's Daily Report for April 12 would not have referenced a 3:30 a.m. inspection by the Coast Guard had such an inspection occurred. See T & T Daily Reports, 4/12/14 at 2 (summarizing daily activities). In so finding, the Court recognizes that the Coast Guard's records are not immune from error: They are the product of reports from the field filtered through one or more offices, such that errors could occur. See generally Tr. 175 (Edgar). However, the lack of a reference in T & T's records-or in any records created during the broader time period, other than the Coast Guard report on which Starr relies-to a 3:30 a.m. inspection is telling.
In any event, ultimately, nothing of substance turns on whether the wing tank damage was first found at 3:30 a.m. by the Coast Guard or at 9:00 a.m. by T & T and the Coast Guard. Starr no longer contends that this damage was caused by the lightering (rather than, as WQIS argues, by the impact of the grounding itself, see Tr. 422-24 (Randall) ). The difference of some five and a half hours is of no practical significance.
3. April 12: The Revised Lightering Plan
By the morning of April 12, the water level in the Mississippi River had dropped to 17.1 feet, which left the barges "high and dry" on an exposed sandbar. JSF ¶ 35. The barges nevertheless remained "in stable condition resting on [a] sandy bottom." T & T Daily Reports, 4/12/14, at 1.
Later that day, after consultation with the Coast Guard, see Tr. 259-66 (Schrader), T & T submitted a revised salvage plan, which the Coast Guard approved the next day, JSF ¶ 34; see also JX D (the "April 12 Salvage Plan"); T & T Daily Reports, 4/13/14 at 2 (reporting Coast Guard's "verbal[ ] approv[al]" of revised plan during meeting). The revised plan "propose[d] an alternative plan that [could] be taken to resume lightering operations." April 12 Salvage Plan at 2. As T & T explained (and as the Court has found), lightering under the April 9 Plan had been halted because, after two and a half hours of lightering in the afternoon of April 11, it had become "evident that the river conditions were deteriorating rapidly and the risk of the support assets become grounded was great." Id. at 2. The April 12 Plan thus called for abandoning "lightering by means of direct barge to barge contact" because of the falling water levels in the river. Id. Instead, the April 12 Plan called for lightering by means of "Over Water Pumping," with the receiving barge positioned farther out into the Mississippi and the "spar barges in deeper water." Id. at 7.
The April 12 Plan reported the "Barges in stable condition resting on sandy bottom." Id. at 3. The Plan noted two concerns: GM-5001 was resting "on irregular bottom support," and in GM-5002, "the starboard # 3 wing tank [was] showing signs of metal failure." Id. The plan nevertheless called for lightering of the barges *560in the same order "as the original plan." Id. at 7.
On April 13 and 14, T & T prepared to implement the revised plan, coordinating the movement of equipment and personnel to the site and the assembly of the barges and pipeline necessary for lightering. See T & T Daily Reports, 4/13/14 at 2, and 4/14/14 at 2. Throughout those two days, T & T continued to report that the barges were in stable condition on a sandy bottom, with the GM-5001 resting on irregular bottom support, and that no sagging, hogging, or torque was observed. See T & T Daily Reports, 4/13/14 at 1, and 4/14/14 at 1.
4. April 15: Lightering Resumes
On April 15, lightering operations restarted. JSF ¶ 37. T & T Salvage again reported that the barges were in stable condition, resting on a sandy bottom. T & T Daily Reports, 4/15/14 at 1. Although the original salvage plan had called for the lightering of the 5002 first-and the 5002 had begun to be lightered on April 11-under the revised salvage plan, the GM-5001 was lightered first. JSF ¶ 37; see T & T Daily Reports, 4/15/14 at 3.
At 7:35 p.m. on April 15, T & T began pumping oil off of the GM-5001. T & T Daily Reports, 4/15/14 at 3. Pumping continued through the next two days, see T & T Daily Reports, 4/16/14 at 3, and 4/17/14 at 3, and lightering of the GM-5001 was completed April 17, JSF ¶ 37. Lightering of the GM-5002 then began on April 17 and was completed April 19. JSF ¶ 37. After the lightering was complete, some quantity of oil nevertheless remained in the barges. JSF ¶ 38.
5. April 17: Refloating after Lightering
On April 17, T & T Salvage submitted a Preliminary Salvage and Refloating Plan to the Coast Guard for the purpose of "discuss[ing] actions to be taken in connection with the refloating of the tank barges following completion of the lightering operation." JX E (the "April 17 Refloating Plan") at 2; see JSF ¶ 39. That plan summarized T & T's earlier assessment of the site, including its conclusion that, although a portion of the hull of the 5001 was unsupported, "the unsupported section did not pose concern for [the] integrity of the hull." April 17 Refloating Plan at 5. T & T also reported that its "frequent[ ]" monitoring and inspection of the barges during the lightering period revealed "[n]o new damage or deterioration ... since the inspection on 4/12/2014." Id. The refloating plan consisted primarily of scouring the sand away from the underside of the barges with high-powered fire-hoses. April 17 Refloating Plan at 12; see also T & T Daily Reports, 4/18/14, at 3.
The Coast Guard approved that plan. JSF ¶ 39. On April 19, with water levels rising, see T & T Daily Reports, 4/19/14 at 1, the refloating operation began, JSF ¶ 41. Water levels, however, did not rise as fast as T & T had anticipated. See T & T Daily Reports 4/20/14 at 1, 6. T & T therefore submitted an Addendum to the Salvage and Refloating Plan. See JX F ("Refloating Plan Addendum"). That Refloating Plan Addendum proposed using an excavator to "displace sand away" from the barges, primarily from the 5001. See id. at 2. The Coast Guard also approved the addendum. JSF ¶ 40. The GM-5002 was refloated on April 26. JSF ¶ 41; see T & T Daily Reports, 4/26/14 at 3. The GM-5002 was refloated on April 30.
E. The Coast Guard's Actions During This Period
As the foregoing chronology suggests, the Coast Guard was involved in the lightering and refloating of the barges throughout much of the salvage operation. The *561facts with respect to much of the Coast Guard's response to the grounding are not in dispute.
Officers from the Coast Guard's MSU-Paducah office first responded to the grounding on April 7, 2014. See Coast Guard Report at 45. Following the grounding, the MSU-Paducah designated Coast Guard Chief Petty Officer Heather Norman as the Federal On-Scene Coordinator Representative (the "FOSCR") for the incident. JSF ¶ 48.9 Although the FOSCR is an "On-Scene" coordinator, Norman was physically present at the site of the incident for only one day during the salvage operation. JSF ¶ 50. For the remainder of the operation, Norman supervised the Coast Guard's response from the Coast Guard's offices in Paducah, Kentucky. See Norman Decl. ¶ 25.
On April 12, the Coast Guard's MSU-Paducah station called for the Coast Guard's Atlantic Strike Team to assist with its supervision and monitoring of the lightering and salvage operation. See Coast Guard Report at 51-52; JSF ¶ 45. "The three Strike Teams (Atlantic, Gulf, and Pacific) provide trained personnel and specialized equipment to assist the [On-Scene Coordinator] in training for spill response, stabilizing and containing the spill, and in monitoring or directing the response actions of the responsible parties and/or contractors." 40 C.F.R. § 300.145(a)(1). As Norman explained, "[h]aving a strike team on site was valuable not only because the team possessed greater expertise in responding to this type of incident, but also because their presence bolstered the manpower of the overstretched small unit whose members were working 12- to 14-hour days." Norman Decl. ¶ 15. On April 14, a four-member team from the Atlantic Strike Team arrived on site. Coast Guard Report at 54; JSF ¶ 46. On April 19, the Strike Team force was reduced from four to two and personnel from MSU-Paducah left the scene. JSF ¶ 46, 47. The Strike Team left the scene altogether on May 2, 2014. See Norman Decl. ¶ 15.
The parties are also largely in agreement as to what the Coast Guard did not do. As the parties have stipulated, the Coast Guard did not issue a Captain of the Port order or set up a formal Incident Command to supervise the lightering and salvage operation.10 See JSF ¶¶ 51, 52. Nor did the Coast Guard issue a Notice of Federal Interest (NOFI) in response to the grounding. Although Officer Norman testified that the Coast Guard's practice is to "issue[ ] a Notice of Federal Interest to the responsible parties, or those suspected of being responsible parties, in the aftermath of an incident that constituted a substantial threat of discharge," see Norman Decl. ¶ 7, the Court finds that the Coast *562Guard did not issue a NOFI in this case. There is no evidence in the record that the Coast Guard issued such a notice. Rather, Norman's testimony is best read to demonstrate that, although issuance of such a notice is standard Coast Guard practice when an incident poses a substantial threat of discharge, no such notice was issued here.11 Had a NOFI been issued, it would have been issued to Genesis and it would have been discoverable in this litigation. See Tr. 300-04 (Russell). No NOFI, however, has been produced-including by Genesis, the insured in whose shoes Starr stands as subrogee-and no representative of either the Coast Guard or Genesis testified that a NOFI ever issued.
The parties' principal dispute as to the Coast Guard's involvement is whether, in real time, the Coast Guard assessed that the barges presented a substantial threat of discharge of oil, and whether such an assessment animated the Coast Guard's actions. Here, too, the Court finds with WQIS. On its review of the evidence and fair inferences therefrom, the Court finds that the Coast Guard did not assess, and its actions were not consistent with or prompted by a real-time assessment, that the barges posed a substantial threat of the discharge of oil. The Court's view is informed by its resolution of two underlying factual disputes.
First, although the parties agree that the Coast Guard activated the National Pollution Fund, they dispute the basis for the Coast Guard's activation of the Fund and the inferences to draw from its activation. See Pl. Br. at 11; Def. Reply Br. at 4-5. The parties agree that the National Pollution Fund may be activated in the event of either an actual oil spill or a substantial threat of discharge. See Def. Reply Br. at 5. The Court finds that the National Pollution Fund was activated on the basis of an erroneous determination that there had been an actual 1-gallon spill of oil. Important here, the Court finds that the Fund was not activated based on any determination that the grounded barges posed a substantial threat of discharge.
An April 12, 2014 email from the National Pollution Fund Center in Washington, D.C. to the Coast Guard's Sector Ohio Valley in Louisville, Kentucky, is significant evidence on this point. The email confirms that that the National Pollution Fund was activated in response to the grounding of the Genesis barges. See Coast Guard Report at 65 (reproducing this email). It summarizes the details of the grounding as then understood by the Coast Guard. Puzzlingly, the email recites the "estimated quantity of oil discharged" as "1 Gallons," id. -even though there is no evidence that any oil whatsoever was discharged in connection with the grounding, and the parties have stipulated that there never was any such discharge, JSF ¶ 42. At the same time, the email addresses the "FOSC determination of substantial threat." Coast Guard Report at 65. FOSC, again, refers to the Coast Guard's designated representative in connection with the incident, Officer Norman. See JSF ¶ 48-49. As to the "FOSC determination of substantial threat," the April 12 email states: "N/A." Coast Guard Report at 65. The April 12 email is contemporaneous evidence that the National Pollution Fund was activated on the basis of an erroneous belief within the Coast Guard that there had been a 1-gallon spill, and not because of a real-time determination by Captain Norman (or any other Coast Guard official) that the grounded barges posed a *563substantial threat of discharge. The record, in fact, is devoid of any written evidence at any point during the period of the barges' salvage reflecting such a determination by the Coast Guard.
Second, the parties disagree whether the Coast Guard expressed concern about the threat of oil discharge during conference calls with on-scene personnel during salvage operations. The Court finds, with WQIS, that, although the Coast Guard participated in daily conference calls with personnel on scene during salvage operations, the evidence as to the content of these calls does not reveal any contemporaneously expressed concern by the Coast Guard that the barges posed a substantial (or any) threat of discharge.
To be sure, Starr has a basis in the trial record for asserting otherwise. As Starr notes, in the deposition on written questions that she submitted in this lawsuit, Norman attested that, on these daily calls, the Coast Guard informed other participants that the Coast Guard believed the barges posed a substantial threat of discharge. See Pl. Br. at 6 (citing Norman Decl. ¶ 16). However, viewing Norman's testimony in light of the circumstances in which it was received-and in light of the other evidence in the case, which uniformly is to the contrary-the Court does not credit that factual assertion of Norman's.
The Court set out the circumstances under which Norman's written testimony was given in its bench ruling receiving, with qualifications, this testimony. See generally 12/4/17 Tr. 2-14. In brief, Starr sought testimony from a Coast Guard witness as, effectively, a Federal Rule of Civil Procedure 30(b)(6) corporate witness. The Coast Guard declined to make a witness available for live testimony, relying on its Tuohy regulations, which permitted the Coast Guard to demur from putting forward personnel for live testimony. See Dkt. 34 (letter from Starr); 6 C.F.R. §§ 5.41 - 5.49 ( Touhy regulations). See generally U.S. ex rel. Touhy v. Ragen , 340 U.S. 462, 71 S.Ct. 416, 95 L.Ed. 417 (1951). The Coast Guard agreed only that Norman would respond in writing to 15 questions from each party-with each party to identify 10 "direct examination" questions to be put to Norman, and, after seeing its adversary's 10 questions, five further "cross examination" questions. See 12/4/17 Tr. at 6. Significantly, all questions were to be formulated and put to Norman before any answers were received from her. There was no opportunity for counsel to follow up in response to Norman's answers. This format did not allow counsel to probe the factual basis for any unsourced proposition to which Norman attested.
The nature of the Norman deposition on written questions prevents the Court from concluding that Norman's testimony with regard to the specific content of the daily conference calls with the other salvage participants is meaningfully reliable and worthy of weight. Inherently, of course, a Rule 30(b)(6) deponent "need not have personal knowledge concerning the matters set out in the deposition notice"; when such a deponent lacks such knowledge, "the corporation is obligated to prepare them so that they may give knowledgeable answers." Agniel v. Cent. Park Boathouse LLC , No. 12 CIV. 7227 NRB, 2015 WL 463971, at *2 (S.D.N.Y. Jan. 26, 2015) (internal quotations omitted). The questions put to Norman thus did not necessarily purport to probe her personal knowledge. Ordinarily, counsel may inquire of the Rule 30(b)(6) corporate witness what the basis was of particular testimony, which affords counsel-and the trier of fact-a basis for gauging the reliability of that testimony (and may enable counsel to follow up with witnesses with personal knowledge). Here, however, given the *564strictures imposed by the Coast Guard, counsel were unable to probe Norman as to the basis for her testimony on particular points where her answers did not make this clear.
In its December 4, 2017 bench decision explaining its decision to receive Norman's testimony, the Court noted that, because Norman's testimony was not subjected to cross and redirect examination, this testimony might prove less reliable and worthy of less weight. 12/4/17 Tr. at 13-14. The Court explained:
Ms. Norman's testimony was not subjected to cross- and redirect examination in the conventional dynamic sense. The questions used on cross examination were promulgated after the other side's direct questions were known but before the answers to them were known. This necessarily inhibited all parties in [boring] down on the Coast Guard testimony, including to understand the precise factual basis for Ms. Norman's statements. The Court has no doubt that, relative to the live testimony, the testimony given under the specifications to which the parties and the Court agreed was less revealing. A live examination would have produced a richer, more nuanced deposition. The Court must and will consider the manner in which Ms. Norman's testimony was elicited as going to its weight. In assessing the appropriate weight, the Court will pay particularly close attention as to any p[articular] answer to the extent to which Ms. Norman's testimony is or is not corroborated either by other witnesses, documents, or circumstantial logic.
12/4/17 Tr. 13-14.
As the Court anticipated in its bench decision, Norman's testimony as to the content of the conference calls merits substantially reduced weight. Norman attested that "the Coast Guard conducted conference calls two or three times a day" with Genesis and the other interested parties and "it informed the parties that, pursuant to the NCP, there was considered a substantial threat of discharge so long as the barges remained grounded with residual oil onboard." Id. (emphasis added). Starr relies on Norman's declaration in contending that the Coast Guard-in real time-believed the barges posed a substantial threat of discharge and communicated that concern to Genesis and T & T. But Norman's declaration on these points is elusive. She does not represent that she has personal knowledge of the calls in question. She does not represent that she was personally present for any call in which the statements in question were made (or that she made such statements). She does not recite the basis for her knowledge or whether she took any steps to refresh her recollection on these points. And, as the words from her testimony as italicized above show, Norman's locutions-including her anthropomorphic identification of participants ("the Coast Guard conducted"), her unspecific identification of speakers ("it informed the parties"), and her use of the passive voice ("the was considered")-leave indeterminate the source and basis of her knowledge. These limitations prevent the Court from crediting Norman's testimony on these points unless corroborated.
All other evidence, however, is to the contrary. The trial record is devoid of any contemporaneous documentation of such an expressed concern by the Coast Guard. And the witness testimony is to the contrary. The Court credits the testimony on this point of Ebanks and Schrader. Each testified that, on the daily conference calls, representatives of the Coast Guard did not describe the barges as posing a substantial threat of discharge. Schrader, who attested *565that he was present for and participated in all of the daily 2 p.m. telephone calls, see Schrader Decl. ¶ 50; Tr. 269 (Schrader), did not hear anyone from the Coast Guard express any concern for the structural integrity of the barges or for the threat of discharge they might have posed, see Tr. 269, 271-72, 277-78 (Schrader). Ebanks' testimony is in accord. Although Ebanks was not on site during the lightering operation, see PX 26 ("Ebanks Dep.") at 29, she testified to participating in the daily 2 p.m. calls with the Coast Guard's Strike Team, id. at 69-70. On those calls, she testified, the Coast Guard "would give a briefing on what was happening actually on site." Id. at 70. But Ebanks did not recall the Coast Guard on any of these calls ever expressing "any concerns about the project or any pollution risk." Id. The testimony of Ebanks to which Starr cites is not to the contrary. Starr notes that, when Ebanks was asked whether the Coast Guard was concerned about the risk of pollution, she replied, "Yes." Ebanks Dep. at 37; see Pl. Br. at 6-7. But when asked whether the Coast Guard told Ebanks (and the other conference call participants) of the Coast Guard's concern, all Ebanks could say was, "Well there was a risk of pollution." Id. That testimony does not supply well-founded support that the Coast Guard communicated any concerns about the risk of a discharge of oil.
Considering all trial evidence, the Court therefore finds that the Coast Guard, in real time, did not perceive a substantial threat of a discharge, articulate to other players that there was such a threat, or take action on the basis of such a threat.
F. Insurance Payments and Settlement Efforts
In the period after the barges were refloated, Genesis made several claims on insurance policies relevant to the grounding of the barges. Genesis also attempted, unsuccessfully, to broker a settlement among the disputant insurers here.
1. Claims by Genesis to WQIS
Genesis, through its agent, McGriff, presented a claim of $210,559.43 for the cost of the Oil Spill Response Organization, SWS Environmental, to WQIS. JSF ¶ 55. WQIS paid that claim on June 3, 2014. Id. WQIS does not dispute that SWS's costs, which substantially involved mandatory measures to protect against spillage during lightering (e.g. , the use of booms) without regard to the degree of risk of such spillage, fell within its policy.12 These costs are not at issue in this litigation.
2. Claims by Genesis to Starr
Genesis, through McGriff, presented a claim to Starr in the amount of $2,892,670.37 for the entire amount invoiced by T & T Salvage to Genesis for the work done to salvage the barges. JSF ¶ 57. Responsibility for the T & T Salvage costs are at the heart of this litigation. Starr contends that these were incurred to mitigate or prevent a substantial threat of discharge and thus are properly chargeable to WQIS.
On October 22, 2014, Starr nevertheless made a "Payment on Account" of $1.5M to Genesis Marine. JSF ¶ 59.
3. Unsuccessful Settlement Efforts13
In an attempt to resolve the dispute among Genesis's insurers, McGriff, on *566Genesis's behalf, retained Merrill Marine to conduct an apportionment of the T & T Salvage invoice between the Starr Indemnity Policy and the WQIS Policy. JSF ¶ 61. Pennington of Merrill Marine performed the allocation. JSF ¶¶ 61-62. He apportioned SWS's charges-the Oil Spill Response Organization onsite-to WQIS and T & T Salvage's bills to Starr. See DX 45. Merrill later modestly revised this allocation to reduce the allocation chargeable to WQIS from $28,043.28 to $27,913.38, with the remainder-$2,864,756.26-allocated to Starr. See DX 46. On November 19, 2014, Merrill submitted this final allocation to Starr and WQIS. JSF ¶ 64. On December 3, 2014, WQIS paid Genesis the amount Merrill had allocated to it. JSF ¶ 65. Starr, however, rejected McGriff's proposed apportionment. JSF ¶ 66.14
On December 16, 2014, McGriff's president sent an email to Starr expressing the view, consistent with Merrill's final allocation, that Starr was responsible for all but $27,913.38 of T & T's expenses. DX 54. Starr then proposed to WQIS an alternative apportionment. See Tr. 96 (Ferguson). Specifically, on December 17, 2014, Ferguson, vice president of Starr Adjustment Services, the Starr affiliate which handles claims for Starr, see Ferguson Aff. ¶ 1, emailed Diamond of WQIS, see PX 11. The email attached a spreadsheet presenting two alternative means of allocating, between Starr and WQIS, T & T's daily bills during the salvage operation. PX 11. The first apportionment was the one by Merrill's Pennington. The second was by Carolyn O'Connor, an employee of Vericlaim who had handled Genesis's claim for Starr. Id. ; see Tr. at 75-76. She allocated all of T & T Salvage's bills from April 8 to April 19-the date on which lightering of the oil was completed-to WQIS, and the balance to Starr. PX 11. Under that apportionment, WQIS would have been responsible for $1,077,413.35 of T & T Salvage's bills. Id. The trial record does not reflect WQIS' acceptance of O'Connor's proposed allocation. Rather, Ferguson testified that WQIS rejected this proposal. Tr. 96 (Ferguson).
4. Subsequent Events
On December 30, 2014, Starr made an additional "Payment on Account" to Genesis for $800,000. JSF ¶ 68.
On February 2, 2015, Genesis and Starr agreed to a Release and Assignment, JSF ¶ 69, under which Genesis agreed to assign to Starr its claims against WQIS, see JX G ("Release Agreement"). As consideration, Starr paid the remainder of Genesis's claim-$1,111,078.52. Id. at 1; see JSF ¶ 70.
On April 6, 2016, the U.S. Coast Guard sent Genesis Marine a bill for the costs, time, and expenses incurred by the Coast Guard. JSF ¶ 54. That bill totaled $57,243.39. Id.
II. Conclusions of Law
A. Jurisdiction and Applicable Law
Although the parties have not put jurisdiction in dispute, the Court has an independent obligation to ensure its subject matter jurisdiction. The Court holds that Starr has properly invoked the Court's admiralty jurisdiction here.
*567"Title 28 U.S.C. § 1333(1) grants federal district courts the power to entertain any civil case of admiralty or maritime jurisdiction. This grant includes jurisdiction over all contracts which relate to the navigation, business, or commerce of the sea." Fireman's Fund Ins. Co. v. Great Am. Ins. Co. of N.Y. 822 F.3d 620, 632 (2d Cir. 2016) (internal quotations, alterations and citations omitted). "The boundaries of admiralty jurisdiction over contracts are conceptual rather than spatial, and defined by the purpose of the jurisdictional grant-to protect maritime commerce." Folksamerica Reinsurance Co. v. Clean Water of N.Y., Inc. , 413 F.3d 307, 311 (2d Cir. 2005). "[W]hether a contract is a maritime one ... 'depends upon the nature and character of the contract,' and the true criterion is whether it has 'reference to maritime service or maritime transactions.' " Norfolk S. Ry. Co. v. Kirby , 543 U.S. 14, 23-24, 125 S.Ct. 385, 160 L.Ed.2d 283 (2004) (ellipsis omitted) (quoting N. Pac. S.S. Co. v. Hall Bros. Marine Ry. & Shipbuilding Co. , 249 U.S. 119, 125, 39 S.Ct. 221, 63 L.Ed. 510 (1919) ). Thus, the Court's "inquiry focuses on 'whether the principal objective of a contract is maritime commerce.' " Fireman's Fund , 822 F.3d at 632 (quoting Kirby , 543 U.S. at 25, 125 S.Ct. 385 ). The "contract's subject matter must be our focal point." Id. (quoting Folksamerica , 413 F.3d at 312 ).
As to insurance contracts, admiralty jurisdiction exists "where the primary or principal objective of the contract is the establishment of 'policies of marine insurance,' " Folksamerica , 413 F.3d at 315 (quoting New England Mut. Marine Ins. Co. v. Dunham , 78 U.S. (11 Wall.) 1, 35, 20 L.Ed. 90 (1870) ), that is, where an insurer "assumes risks which are marine risks,' " id. at 316 (quoting Jeffcott v. Aetna Ins. Co. , 129 F.2d 582, 584 (2d Cir. 1942) ); see Fireman's Fund , 822 F.3d at 632. "[A]n insurance policy's predominant purpose, as measured by the dimensions of the contingency insured against and the risk assumed, determines the nature of the insurance." Folksamerica , 413 F.3d at 317 (quoting Acadia Ins. Co. v. McNeil , 116 F.3d 599, 603 (1st Cir. 1997) ). " 'Ultimately, coverage determines whether a policy is 'marine insurance,' and coverage is a function of the terms of the insurance contract and the nature of the business insured.' " Fireman's Fund , 822 F.3d at 632 (quoting Folksamerica , 413 F.3d at 316 ).
To determine whether a contract dispute falls within this Court's admiralty jurisdiction, a court, before "inquiring into the subject matter of the contract, [must] first make a 'threshold inquiry' into the subject matter of the dispute ." Fireman's Fund , 822 F.3d at 634 (quoting Folksamerica , 413 F.3d at 312 ). " '[A] federal court must initially determine whether the subject matter of the dispute is so attenuated from the business of maritime commerce that it does not implicate the concerns underlying admiralty and maritime jurisdiction.' " Id. (quoting Folksamerica , 413 F.3d at 312 ).15
Here, the nature of the contract and the nature of the dispute both squarely implicate the Court's admiralty jurisdiction. The operative insurance agreement between WQIS and Genesis provides pollution liability insurance for Genesis's barges and other vessels; and the parties' dispute turns on the applicability of the policy to liability arising out of a marine incident on a vital waterway for U.S. shipping. See generally *5681 Thomas J. Schoenbaum, Admiralty and Maritime Law § 3-10, Admiralty contract jurisdiction (5th ed.) (West) (last visited Apr. 24, 2018).
That this case falls within the "court's admiralty jurisdiction ha[s] implications beyond conferring federal jurisdiction. In particular, when a contract is a maritime one, and the dispute is not inherently local, federal law controls the contract interpretation." Fireman's Fund , 822 F.3d at 632-33 (internal quotations and citations omitted). However, despite the "great importance" of "uniformity ... in admiralty law," State Trading Corp. of India v. Assuranceforeningen Skuld , 921 F.2d 409, 414 (2d Cir. 1990), maritime insurance contracts are subject to an exception to the general federal common law of admiralty, see Wilburn Boat Co. v. Fireman's Fund Ins. Co. , 348 U.S. 310, 75 S.Ct. 368, 99 L.Ed. 337 (1955). Under Wilburn Boat , state law applies to maritime insurance contracts "in the absence of a controlling federal statute or rule." State Trading Corp. , 921 F.2d at 414 (discussing Wilburn Boat ); see also 2 Schoenbaum, supra , § 19-6, Applicable law. The Court's construction and application of the WQIS policy, therefore, is guided by New York state law except to the extent federal law conflicts.
B. The Definition of a "Substantial Threat of Discharge"
As reviewed above, WQIS's liability to Genesis, and thus to Starr, under the WQIS insurance policy turns on three related provisions of the policy. Central to each policy provision is the concept of a "substantial threat of discharge of oil." One provision (subsection A(1) ) turns on whether the grounded barges posed a substantial threat of discharge of oil. A second provision (subsection A(7)(A) ) turns on whether actions taken in connection with salvage were taken for the purpose of mitigating such a substantial threat. The third provision (subsection A(7)(D) ) turns on whether the Coast Guard ordered the lightering of the barges to guard against a perceived substantial threat, or whether, had the lightering not independently occurred, the Coast Guard, on account of such a threat, would have ordered it.
The Court's analysis therefore begins with the following threshold definitional question: For purposes of applying the WQIS Policy provisions at issue, what constitutes a "substantial threat of discharge"?16
Neither the WQIS Policy nor OPA, which subsection A(1) of the Policy expressly incorporates by reference, defines "substantial threat of discharge." The parties have not directed the Court to any case law interpreting the term, nor has the Court found any instructive decisions.17
Starr urges the Court to rely on a regulatory definition the Coast Guard has adopted in its regulations under OPA governing tank-vessel response plans for oil.
*569See 33 C.F.R. § 155.1020. There, the Coast Guard has defined a "substantial threat of discharge" to include "any incident involving a vessel that may create a significant risk of discharge of cargo oil. Such incidents include, but are not limited to, groundings, strandings, collisions, hull damage, fire, explosion, loss of propulsion, flooding, on-deck spills, or other similar occurrences."Id. Starr argues that this definition establishes coverage here, because the events involving the two Genesis barges undisputedly entailed groundings (and also strandings and loss of propulsion).
In construing the similar WQIS policy term, the Court has considered the Coast Guard's regulatory definition under OPA, as WQIS agrees is appropriate. See Def. Pre-Tr. Br. at 4. But that definition, properly read, does not connote that every instance of every occurrence that the Coast Guard lists (e.g., every instance of "flooding," "fires," "hull damage," "on-deck spills," "loss of propulsion," or "groundings") necessarily presents a "substantial threat of discharge." On the contrary, the Coast Guard definition embeds a vital qualification: an incident within the enumerated categories of incidents is a "substantial threat" if it also "create[s] a significant risk of discharge of cargo oil." 33 C.F.R. § 155.1020. Starr reads that limiting clause out of the regulation, enabling the argument that every incident of "grounding, stranding, collision, hull damage, fire, explosion, loss of propulsion, flooding, on-deck spill, or other similar occurrence" is inherently a "substantial threat of discharge." Properly read, the Coast Guard's definition requires that an incident create a significant risk of discharge of cargo oil to constitute a substantial threat. Otherwise, any quotidian flooding (even from a clogged sink or lavatory) or on-deck spill (even of water from a bucket) would, improbably, create per se a substantial threat of discharge of cargo oil.
In the Court's view, there is no meaningful distinction between the Coast Guard's regulatory definition-properly construed-and an interpretation based on the text of the provisions at issue in WQIS's policy. And as a matter of contract interpretation-as with statutory construction-the Court begins with the plain text of the term at issue. See, e.g. , Ment Bros. Iron Works Co. v. Interstate Fire & Cas. Co. , 702 F.3d 118, 122 (2d Cir. 2012) (interpretation of insurance contracts under New York law); United States v. Am. Soc. of Composers, Authors, Publishers , 627 F.3d 64, 72 (2d Cir. 2010) (statutory construction).
As a matter of plain language, the terms "substantial threat" and "significant risk" of discharge each denote a likelihood or probability. That such probability be "significant" or "substantial," in turns, conveys that it be "considerable in quantity, significantly great," see Merriam-Webster, "Substantial," https://www.merriam-webster.com/dictionary/substantial, or "of a noticeably or measurably large amount," see id. , "Significant," https://www.merriamwebster.com/dictionary/significant.
A final point as to plain language: In the WQIS policy and the Coast Guard regulations under OPA, respectively, the word "substantial" qualifies the word "threat" and the word "significant" qualifies the word "risk." These adjectives do not modify the words "discharge" or "oil." Thus, the risk or threat must be substantial or significant; the definition is not met by a trifling threat or risk of discharge of a body of oil, even if, in quantity, the body of such oil is large. To count as "substantial threat of discharge" under the WQIS policy, there must have been a substantial-that is, a considerable, significantly great, noticeably or measurably large-threat of *570discharge. It is that threat (or risk ) that must be significant.
C. Application to the Three Policy Provisions
With that understanding of the term "substantial threat of discharge" in mind, the Court now applies the three provisions of WQIS's policy to the facts found. The Court holds that none of these policy provisions apply.
1. WQIS Has No Liability Under Subsection A(1)
As noted, subsection A(1) of the WQIS Policy insured Genesis for costs and expenses incurred by Genesis for removal under OPA and damages under Subchapter I of OPA for which liability would have been imposed had Genesis not incurred such liability voluntarily. See WQIS Policy at 4. Subchapter I of OPA, in turn, makes "each responsible party for a vessel" liable for specified "removal costs and damages" that result from an "incident" in which "oil is discharged" or which "poses the substantial threat of a discharge of oil." 33 U.S.C. § 2702(a). Those costs include removal costs incurred by the United States, a State, or a tribe, or any costs incurred by a private actor for acts taken consistent with the National Contingency Plan. Id. § 2702(b)(1)(A)-(B). "Removal costs," in turn, "means the costs of removal that are incurred after an actual discharge of oil has occurred or, in a case in which there is a substantial threat of a discharge of oil, the costs to prevent, minimize, or mitigate oil pollution from such an incident." Id. § 2701(31). OPA's reach extends (at a minimum), to waters of the United States that are actually navigable, including, of course, as relevant here, the Mississippi River. Id. § 2702(a) ; see, e.g. , In re Needham , 354 F.3d 340, 344 (5th Cir. 2003). Thus, Subsection A(1) of the WQIS policy insured Genesis for liability imposed under OPA and for costs incurred by Genesis in response to a substantial threat of discharge of oil.
The Court holds that this policy provision does not cover the costs at issue here because (1) no liability has been imposed under OPA, and (2) had Genesis not voluntarily acted to salvage the barges, liability would not have been imposed under OPA.
a. Liability has not been imposed under OPA
Liability has not been imposed under OPA. And Starr, for the most part, does not contend otherwise. See Pl. Br. at 34-35.
A late development, however, warrants elaboration on this point. On January 25, 2018, after trial concluded, the Coast Guard sent WQIS a letter requesting reimbursement for the $57,243.39 in costs the Coast Guard had incurred during its participation in the salvage of Genesis's barges. See Letter from William D. Adkins, U.S. Coast Guard, to Harry J. Diamond, WQIS (January 25, 2018), JX H.18 The letter seeks reimbursement based on the costs the Coast Guard incurred in responding to the grounded barges. The Coast Guard's letter describes the stranded barges as having "posed a substantial threat to discharge oil." Id.19 The record is *571silent as to why the Coast Guard sought such reimbursement from WQIS nearly four years after the fact, and immediately after the trial in this case. It is also silent as to the factual basis, if any, on which Adkins, whom the letter describes a case officer in the Coast Guard's National Pollution Fund Center, came to state that the barges posted a substantial threat to discharge oil.20 Based on this letter, Starr argues that, even if liability under OPA had not previously been imposed, it now has been. See Pl. Reply. Br. at 12.
For the reasons reviewed below, the Court finds, on the basis of a careful consideration of the entire trial record, that the barges did not pose a substantial threat of discharge. The Court thus rejects as inaccurate Adkins's conclusory statement to the contrary. The Court also rejects Starr's argument that the Coast Guard's post-trial demand letter "imposes" liability. On the basis of the record, which is now closed, the letter merely demands payment. WQIS may or may not acquiesce to this demand, and further developments may-or may not-ensue as to the Coast Guard's bid to reclaim the $57,243.39 it expended. But on the basis of the record, no liability has been "imposed" at all.
Equally important, even if the Coast Guard's demand letter were treated as "impos[ing] liability" under OPA, it would do so only as to the costs the Coast Guard incurred for which it now seeks recompense. Those are not the costs at issue here. The costs at issue here are those that Genesis incurred and which Starr, in Genesis's shoes, seeks in this trial to recoup. Subsection A(1) does not provide for coverage by WQIS of all of Genesis's costs in the event that the Coast Guard imposes some liability under OPA. The provision's wording is more precise: It affords insurance coverage to Genesis to the extent liability for removal is actually imposed (or would have been imposed) on Genesis under OPA. As to the expenses incurred by Genesis-the approximately $2.8 million which Genesis paid T & T for its salvage work-liability was never "imposed." These costs are unaddressed by the Coast Guard's recent letter.
b. There was no basis for OPA liability to have been imposed
Nor would Genesis have incurred OPA liability had it not acted voluntarily to salvage the barges. This issue turns on whether the barges in fact posed a substantial threat of discharge.21 Starr claims that the barges did so; WQIS argues that they did not. On the basis of the assembled record-including the expert witnesses; the contemporaneous reports of T & T; and the real-time actions and statements of the participants in the salvage process and its oversight, including Genesis, T & T, and the Coast Guard-the Court holds with WQIS. The salvage and lightering was undertaken here to liberate the stranded barges. The evidence does not reflect-and the participants in real time *572did not conclude-that there was a substantial risk of discharge. Rather, the evidence overwhelmingly showed that the barges-by their nature and in the circumstances at hand-were never at risk (or anywhere close) of the type of failure that might have resulted in an oil discharge.
In reaching this central conclusion, the Court has considered all pertinent evidence. Particularly influential was the following.
Expert testimony : The Court found highly persuasive, and credits in full, the expert testimony of George Randall, called by WQIS. Randall is an MIT-trained naval architect and marine engineer with more than 40 years' experience as a marine surveyor, who has testified as an expert 20 times over the past two decades and provided dozens more expert opinions on matters relating to the damage and salvage of barges and other craft. See DX 65 ("Randall Expert Report"), at Tab 2. He presented at trial as thoughtful, well-informed, objective, clinical, and articulate. See generally Tr. 381-91, 400-60. Based on his review, Randall found no substantial threat of a discharge.
Among other things, Randall put in helpful context the limited evidence of damage to the barges, which damage was a central basis for Starr's argument that there was a substantial threat of discharge. See, e.g. , Pl. Br. at 14. Randall concluded that the "only hull damage or distortion noted [on the barges] was local damage to several [hundred]22 square feet of hull bottom plating and internals of GM5002 in way of [the] No.3 Starboard wing tank." Randall Expert Report at 2. This damage, he concluded, "likely was caused by the initial grounding, although possibly it was caused by uneven local support by the ground as the water receded. In either case, there was very little danger of this initial damage worsening, nor of further damage occurring, because of the "static nature of the loads on the barge while it lay stranded." Id. Randall based his conclusions on the notes Pennington took contemporaneous with the grounding and salvage operations, Randall Expert Report at 3, water-level records from two measuring sites near the grounding, id. , and the T & T Daily Reports, see id. at 5. Randall's report focused closely on the observable "deflections and distortions in the hull" of the barges, id. at 5, and the conditions of the barges' grounding and stranding, id. at 6-7.
More generally, Randall persuasively opined as to the inherent structural soundness of the two barges, and why, while stranded, they did not present a meaningful risk of fuel discharge. The barges were particularly strong-indeed, "among the strongest, and the least likely to suffer hull failure during a stranding." Id. at 4. Their strength derived from several features inherent in their design. They are tank barges with "continuously plated upper decks"-i. e. , they benefit from the added strength and rigidity provided by fully decked cargo holds. Id. As Randall explained, this added strength meant that the barges had "greater resistance to overall bending, for example due to unsupported overhang during a stranding." Id. And the barges were particularly strong relative to other tank barges because they had "raised trunk decks," which provided added longitudinal strength. Id. Most important, the barges were double hulled. Id. The double-hull structure provided two important benefits relevant here: the inner *573hull braced the outer hull, helping the barges "withstand concentrated ground loads" on an uneven surface, and the inner hull provided a second layer of protection in the event of a breach of the outer hull (which, of course, did not occur here). Id.
Randall's report also persuasively opined as to why the condition in which the barges were stranded was particularly unlikely to pose a threat to the barges' integrity. As Randall's report explains, the barges' steel hulls are ductile, meaning they will "stretch and deform substantially before breaking," and can be in use for years with deformities of the kind the barges suffered here. Id. at 6. Minor deformities are of little concern, Randall's report explains, because steel fractures only when under cyclical stress-that is, when it is repeatedly stressed and unstressed in the same way. Id. But once the barges were stranded, they were at "virtually zero" risk of such cyclical stresses. Id. Thus, because the stranding presented "the nearly complete absence of a 'fatiguing' environment, the steel structure of the barges could reliably be expected to suffer no further deterioration" beyond whatever damage was caused by the initial stranding. Id. at 7.
The competing expert report of Kenneth Edgar, offered by Starr, does not persuasively undermine Randall's determinations. See PX 17 ("Edgar Expert Report"). To be sure, Edgar concluded that the barges posed a substantial threat of discharge. But that conclusion was based in significant part on Edgar's reading of the Coast Guard's regulatory definition ( 33 C.F.R. § 155.1020 ) to provide that any grounding presents a "substantial threat of discharge." See Edgar Expert Report at 8.23 The Court has rejected that construction of the Coast Guard regulations. See § II.B, supra . And Edgar's construction of the text of this regulation-a legal determination-is not entitled to weight.24
As to the balance of Edgar's analysis, the Court does not find persuasive his assessment that the barges in fact posed a substantial threat of discharge. Edgar opined that the barges were under considerable stress-and thus at a risk of discharge-based on the assumption that the barges were unsupported by the sand bar on which they grounded. See Edgar Expert Report at 20-21. Edgar's premise was that the structural integrity of the barges while stranded was sufficiently compromised such that the barges were in jeopardy of breaking apart or collapsing in a manner that could lead to a discharge of fuel. Id. at 21. But the evidence does not bear out that assumption. Taken together, the trial evidence (documentary, pictorial, and testimonial) instead supports that the long and low barges were solidly constructed, that their structural integrity was not compromised while resting on the sandbar, and that the barges were supported, in substantial part, by the sand bar itself. See, e.g. , PX 10 at 1; see also Randall Expert Report at 4, 6-9.
Starr separately faults Randall's expert report for relying on "an after-the-fact analysis of the actual damages found in the barges after they were refloated." PI. Reply Br. at 6. Such post-hoc analysis is not *574to be credited, Starr argues, because T & T and the Coast Guard were forced to make their judgments in real-time, without the benefit of knowing whether the damage in fact threatened the integrity of the barges. The evidence, however, does not support that there was a substantial threat of discharge (actual or perceived) at any time: before or during the salvage operations, or viewing the barges' condition and other evidence in hindsight. As an expert hired in connection with this litigation, Randall, like Edgar, necessarily engaged in some post-hoc analysis. Randall's analysis confirms that the real-time assessments by the salvage participants, which are devoid of articulated concerns about a threat discharge, were correct.
T & T's daily reports : The Court also found highly revealing T & T's detailed daily reports. See, e.g. , T & T Daily Reports 4/9/14-4/24/14. These reports by Genesis's salvor are the closest evidence as exists to a contemporaneous diary of the salvage and lightering operation. They record central participants' observations and assessments of the barges. On every day on which T & T was on-site, T & T's reports describe the barges in stable condition, and as resting either on a sandy or muddy bottom. See, e.g. , T & T Daily Report 4/9/14 at 1; T & T Daily Report 4/12/14 at 1; T & T Daily Report 4/14/14 at 1. None of these reports evince, even indirectly, concern on the part of T & T, the Coast Guard, Genesis, or any other person or entity for the structural integrity of the barges. T & T's reports also do not reflect reports of sagging, hogging, or torque in the barges, or other potential indicators of structural infirmity (aside from, after April 12, 2014, the damage to the No. 3 Wing Tank on the GM-5002). Pennington's contemporaneous records are in accord. See DX 3 at 4.
T & T's conduct of the lightering operation : The Court also found probative the manner in which T & T conducted the lightering operation. The revised salvage plan-developed after lightering halted on April 11 and approved by the Coast Guard on April 13-does not purport to address any concerns about the structural integrity of the barges, let alone the concerns that Starr theorizes developed early in the lightering process. See April 12 Salvage Plan at 3; see also Tr. 207 (Edgar) ("Q. But, sir, they didn't change the lightering plan to address any structural concerns, did they? A. No, sir."). Had T & T harbored real concerns about potential discharge of fuel into the Mississippi River, it is likely that its revised lightering plan would have addressed them. Notably, too, the revised lightering plan did not contemplate any change in the order in which tanks were to be lightered. See April 12 Salvage Plan at 7. It also did not anticipate any shifting of cargo within the barges or between them. Id. And under T & T's revised plan, the GM-5001 was lightered before the GM-5002, see T & T Daily Reports, 4/15/14, at 3, even though the GM-5002 had experienced the damage to Wing Tank No. 3 on which Starr has seized in this litigation. Had T & T been concerned that that damage threatened the integrity of the GM-5002, it is likely that the revised plan would have prioritized the removal of oil from the GM-5002. That T & T did not do so tends to undermine Starr's theory that that barge was then perceived as presenting a risk of discharge.
The Coast Guard's deeds and words : As reviewed above, see § I.E, supra , the Court has found that the Coast Guard's actions did not bespeak a real-time assessment of a substantial threat of discharge. Quite the contrary: The Court has found it more probable than not that the Coast Guard did not issue a NOFI, the notice the *575agency is duty-bound to issue "to the responsible parties, or those suspected of being responsible parties, in the aftermath of an incident that constituted a substantial threat of discharge." Norman Decl. ¶ 7. And, as the parties have stipulated, the Coast Guard did not issue a Captain of the Port order or set up an Incident Command to supervise the lightering and salvage operation. See JSF ¶¶ 51, 52. Further, as the Court has found, the best evidence is that the Coast Guard, in participating in the daily 2 p.m. conference calls, never expressed concern as to a threat (let alone substantial) of discharge. Viewed together, the Coast Guard's actions (and lack of action) reveal an agency monitoring a salvage operation in an important commercial waterway, not an agency responding to an incident that it viewed as presenting a serious threat of an oil discharge.
The Court therefore finds that, had Genesis not acted on its own to undertake salvage operations, it would not have incurred OPA liability. Subsection A(1) of the WQIS policy insured Genesis for OPA liability where it was responsible for removal costs and damages resulting from a vessel presenting a substantial threat of discharge of oil, 33 U.S.C. § 2702(a), but the two stranded Genesis barges never posed such a threat.
Starr, finally, contends that WQIS had a "tradition and/or course of conduct" of providing coverage under subsection A(1) for costs of this kind. See Pl. Br. at 21-22. The factual record establishes considerably less. WQIS' Vice President of Claims, Larry Diamond, testified that WQIS, for business or client-relations reasons, from time to time has paid claimants in excess of, or without regard for the strict limits of, it policy liability. See Tr. 314-19, 334-36 (Diamond). That included, he testified, the costs incurred by SWS in conjunction with this incident. Tr. 314, 325-32 (Diamond). WQIS's business decision along these lines does not bear on the construction of its policy or its application to the facts adduced at trial. It does not disturb the Court's finding that WQIS was not liable to Genesis under subsection A(1) for the salvage costs at issue.
2. WQIS Has No Liability Under Subsection A(7)(a) of its Policy
Under Subsection A(7)(a), WQIS insured Genesis for
Costs and expenses incurred by [Genesis] for firefighting, salvage or removal of wreck or debris of any Vessel(s) or cargo carried aboard any such Vessel(s), to the extent that such actions were undertaken for the purpose of stopping a discharge or release, or mitigating or preventing a substantial threat of a discharge under OPA ....
WQIS Policy at 5.
For the reasons reviewed above, the Court finds that there was not a substantial threat of discharge. That, however, does not dispose of Starr's claim under subsection A(7)(a), because Genesis, in theory, might nevertheless have incurred expenses to mitigate or prevent a threat of discharge that it mistakenly viewed as substantial. Expenses thus undertaken could have been incurred "for the purpose of ... mitigating or preventing a substantial threat of discharge." The Court, however, finds that Genesis and its agents did not lighter the barges for that purpose-to mitigate even a perceived threat of discharge.
At the outset, it is important to construe this provision. One disputed issue concerns the point in time at which Starr's purpose is properly assessed under this subsection. WQIS argues that the relevant point is the time at which the expenses in question-the lightering or other salvage expenses-were *576incurred. Starr takes the contrary view. It argues that the "purpose" relevant under this provision may have come to exist only after the lightering. Starr further argues that the "purpose" of parties other than those who caused the activity at issue (e.g. , the lightering) to occur may also control. On this basis, Starr suggests that an unexpressed-or even after-the-fact-perception by the Coast Guard of a substantial threat of discharge may trigger this subsection. See Pl. Br. at 35-37.
Starr's reading of subsection A(7)(a) is unpersuasive. The text of that provision makes clear, as WQIS argues, that the purpose must have existed in real time, at the time of the expense at issue, and that the proper focus is on the actor who caused the expense to be incurred. The subsection provides coverage for salvage and removal costs "to the extent that such actions were undertaken for the purpose of ... mitigating or preventing a substantial threat of a discharge under OPA." It follows that coverage under this provision is defined by the reasons that motivated the actor or actors who undertook the "actions" at issue. The "purpose," therefore, must be that of the person or entity that "undert[oo]k," or caused these actions-here, the lightering. Had the Coast Guard ordered the barges' lightering, the actor whose purpose would control would be the Coast Guard. But here, Genesis, and its salvor, T & T, not the Coast Guard, ordered and directed the lightering. WQIS is clearly correct that the " 'purpose' for one's actions in real time cannot be said to change at a later point in time, due to later discovered facts." Def. Br. at 39.
A second issue of construction that separates the parties concerns the possibility of a dual "purpose." WQIS argues that salvage costs are covered by this provision only if the costs were incurred for the exclusive purpose of preventing a substantial threat of fuel discharge. See Def. Br. at 10-12. In so arguing, WQIS relies largely on the Ninth Circuit's decision in Port of Portland v. Water Quality Insurance Syndicate , 796 F.2d 1188, 1195 (9th Cir. 1986) and the District of New Jersey's decision in Kearny Barge Co., Inc. v. Global Insurance Co. , 943 F.Supp. 441, 458 (D. N.J. 1996). Starr takes the opposite view. Starr further argues that the "discharge" purpose may be met even if a real or perceived threat of discharge did not motivate the decision to lighter, so long as the lightering process was made more costly because it was designed to reduce the risk of oil spillage during the lightering. Pl. R. Br. at 4.
As a factual matter, Starr is correct that a portion of the lightering costs was incurred because the cargo of Genesis barges consisted of oil (as opposed to, say, species of cargo that were simpler to remove), because removing oil required the salvor to use procedures aimed at assuring that the lightering process itself did not result in spillage of fuel. But, under the text of subsection A(7)(a), that fact-that the cost of lightering was increased to mitigate the risk of pollution during lightering-does not control. The decisive question instead is whether Genesis incurred costs for the purpose of mitigating or preventing a substantial threat of a discharge under OPA. The Court has held above that it did not.25
*577For this reason, the Court has no occasion to resolve the construction issue over which the parties spar: whether, to trigger coverage under subsection A(7)(a), Genesis's purpose must be exclusively or only partly to address a substantial threat of discharge. Simply put, the Court finds that Genesis and salvor T & T were not motivated at all by an actual or perceived substantial threat of discharge. The Court has earlier recounted the evidentiary basis for its findings that the barges did not pose a substantial threat of discharge and that neither T & T nor Genesis believed there to be a substantial threat of discharge.
Significant, too, the record reflects sound alternative reasons to remove the oil cargo aboard the barges: to lighten and thereby free the barges and to enable the oil to be moved downriver for sale. Removing the weighty oil made the task of dislodging the barges from the sandbar substantially easier. See, e.g. , Norman Decl. ¶ 13. All agreed that removing the oil was necessary-just as it would have been for any cargo-to lighten the barges sufficiently to dislodge them from the strand. See, e.g. , Coast Guard Report at 47 (describing failed attempts to dislodge barges without lightering); April 9 Salvage Plan at 3 (same). As T & T explained, the falling water levels meant that the barges would not become unstuck on their own; rather "the barges ... require[ed] lightering and external intervention." April 9 Salvage Plan at 3.
There were also sound economic reasons to want the oil removed. The oil aboard the barges was valued at nearly $2.5 million, see DX 55 at 1180 (calculating value of cargo on board GM-5001 and GM-5002), and was the property of Genesis's customer, BP, see Tr. at 258-59 (Schrader). Genesis's and T & T's decision to remove the fuel thus is readily explained by reasons other than concern about a fuel discharge.
3. WQIS Has No Liability Under Subsection A(7)(d) of its Policy
The Court, finally, addresses subsection A(7)(d) of its policy. It provided coverage-in the event of an oil discharge or a substantial threat of same-for costs Genesis incurred "with the prior consent of WQIS" and "at the order of the U.S. Coast Guard or designated Federal On-Scene Coordinator." WQIS Policy at 5.26
It is undisputed that the Coast Guard (including through its FOSCR, Norman) did not order Genesis to remove the oil from its barges. As noted earlier, however, the parties dispute whether the Coast Guard would have ordered Genesis to remove the oil, had Genesis not acted on its own.27 Starr relies on Norman's claim in *578her written testimony that the Coast Guard believed that Genesis was "responding properly" to the grounding. See Pl. Br. at 43 (citing Norman Decl. ¶¶ 7, 9, 14, 20). Starr derives from this the inference that, had Genesis not been responsive, the Coast Guard would have ordered the lightering pursuant to its authority under OPA.
In light of the Court's findings of fact, above, this theory is quickly interred. The Court has found that the barges did not pose a substantial threat of discharge. The Court has further found-as reviewed earlier in its discussion of Officer Norman's testimony-that the Coast Guard did not believe, in real time, that the barges posed such a threat. In light of these findings, the Court rejects as factually unfounded Starr's assertion that the Coast Guard would have ordered the lightering of the barges as a response to such a threat.
CONCLUSION
For the reasons reviewed above, the Court holds that WQIS is not liable for the salvage costs incurred by Genesis under any of the three policy provisions at issue, and thus is not liable to Starr.
The Clerk of Court is respectfully directed to enter judgment for WQIS and to close this case.
SO ORDERED.