ORDER

Lennar Mare Island, LLC (LMI), CH2M Hill Constructors, Inc. (CCI), and Steadfast Insurance Company dispute their obligations with respect to the cleanup of Mare Island, a former U.S. Navy shipyard. This order addresses LMI’s motions for partial summary judgment of the definitions of “Government Authority” and “Known Pollution Conditions.” ECF Nos. 160, 186. The court held a hearing on August 7, 2015. Ryan Werner appeared for LMI, Deborah Ballati and Amanda Hair-ston appeared for CCI 1 , and Dale Oliver and John Purcell appeared for Steadfast. The court allowed post-hearing briefing, now on file. The motions are GRANTED IN PART as set forth in this order.

1. UNDISPUTED FACTS

The parties do not dispute this case’s general history, which the court has summarized in previous orders. See Order Feb. 28, 2014, at 6-9, ECF No. 95; Order May 15, 2014, at 2-4, ECF No. Ill; 2 Order Apr. 7, 2015, at 2-3, ECF No. 264. 3 The claims here stem from environmental clean-up work LMI and CCI have undertaken at the former U.S. Navy shipyard on Mare Island in Vallejo, California. LMI agreed to clean up the former base and contracted with CCI, who would investigate and remediate pollution. LMI is also a party to a consent agreement with the City of Vallejo and the California Department of Toxic Substances Control (DTSC). Steadfast Resp. Stmt. Undisputed Material Fact re Gov’t Auth. (Auth. UMF) no. 7, ECF No. 170. In short, this consent agreement governs LMI’s clean-up efforts at Mare Island. See id. nos. 8-15.

At about the time LMI and CCI agreed to undertake efforts to clean up the base, *953 Steadfast issued two insurance policies to LMI and CCI: (1) the Remediation Stop Loss or RSL policy, now expired, which provided coverage to CCI should the cost of the cleanup of “Known Pollution Conditions” exceed a specific amount; and (2) the Environmental Liability Insurance or ELI policy, which among other things provides coverage to LMI for the cost of remediating pollution other than the Known Pollution Conditions. Among other limitations, the ELI policy covers clean-up costs only if (1) “required by Governmental Authority,” Auth. UMF no. 3, 4 and (2) costs are not expended on the remediation of “Known Pollution Conditions.” Auth. Evid. Ex. 1, at SJCE 167596.

LMI alleges Steadfast caused it several million dollars in damages by refusing to pay or delaying payments for claims under the ELI policy. First Am. Compl., ECF No. 22. The complaint describes seven environmental clean-up sites in particular. See id. ¶34. LMI also seeks declaratory judgment that Steadfast must pay specific claims. See id. ¶¶ 34-35. CCI alleges similarly that Steadfast wrongfully withheld payments under the RSL policy and seeks damages and declaratory relief. Answer & Countercl., ECF No. 12. Steadfast denies these allegations and asserts counterclaims against both LMI and CCI. After a series of dismissals and amendments, Steadfast now proceeds on contract claims and seeks a declaration of its rights under both policies. See generally Order Mar. 3, 2016, ECF No. 351; 5 Steadfast’s Second Am. Countercl., ECF No. 315.

This order addresses two motions for summary judgment LMI filed in December 2014 and January 2015. The first motion seeks an order that the DTSC consent agreement is “Government Authority” and requires LMI to clean up any and all contamination it discovers at the seven Mare Island sites. Mem. P. & A. Summ. J. Gov’t Authority (Auth. Mem.) 8, ECF No. 160-1. In its reply, LMI tempers this request and asks only for an order specifying the DTSC agreement “constitutes Governmental Authority” and “if LMI incurs a cost in complying with the Consent Agreement, it is a cost required by Governmental Authority within the meaning of the ELI Policy.” Reply Gov’t Auth. 3, 8, ECF No. 226.

In its second motion, LMI seeks an order defining “Known Pollution Conditions” to be only those conditions listed in certain tables and figures attached to the RSL policy. Mem. P.&A. Summ. J. Pollution (Poll. Mem.) 2, ECF No. 187. The terms for which LMI seeks definitions are found in the ELI and RSL policies and in the endorsements and other materials attached to those policies.. The parties agree those documents govern their dispute. See Auth. UMF nos. 1, 3, 6-15; Resp. Statement Undisputed Material Facts re *954 Known Pollution Condition (Poll. UMF) nos. 1,2,4-9, ECF No. 195.

Steadfast opposes both motions, Auth. Opp’n, ECF No. 169; Poll. Opp’n, ECF No. 193. In general, it contends the ELI policy terms are not susceptible to the definitions LMI proposes, and in any event, if they are, Steadfast argues the terms must be interpreted with the benefit of extrinsic evidence, CCI takes no position on these definitions or LMI’s pending motions for summary judgment.

After the parties presented their arguments at hearing, the court granted Steadfast and LMI leave to file supplemental briefing addressing two specific questions: (1) “Under California law, is an ambiguous contract term a prerequisite to admission of extrinsic evidence of the contracting parties’ course of performance?”; and (2) “Where in the briefing on file, if at all, has Steadfast cited to portions of the record to show there is a genuine dispute of material fact whether the parties’ course of performance explained or supplemented the ELI policy’s definition of ’Known Pollution Condition’?” Order Aug. 14, 2015, ECF No. 289. Both parties submitted responsive supplemental briefs. Suppl. Opp’n, ECF No. 292; Suppl. Reply, ECF No. 299.

II. LEGAL STANDARD

A court must grant a motion for summary judgment “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). A motion for summary judgment calls for a “threshold inquiry” into whether a trial is necessary at all, that is, whether “any genuine factual issues ... properly can be resolved only by a finder of fact because they may reasonably be resolved in favor of either party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250 , 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986). The court does not weigh evidence or assess the credibility of witnesses; rather, it determines which facts the parties do not dispute,’ then draws all' inferences and views all evidence in the light most favorable to' the nonmoving party. See id. at 255 , 106 S.Ct. 2505 ; Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587-88 , 106 S.Ct. 1348 , 89 L.Ed.2d 538 (1986). ‘Where the record taken as a whole could not lead a rational trier of fact to find for the non-moving party, there is no ’genuine issue for trial.’” Matsushita, 475 U.S. at 587 , 106 S.Ct. 1348 (quoting First Nat’l Bank of Ariz. v. Cities Serv. Co., 391 U.S, 253, 289, 88 S.Ct. 1575 , 20 L.Ed.2d 569 (1968)).

The moving party bears the initial burden of “informing the district court of the basis for its motion, and identifying those portions of the [record] which it believes demonstrate the absence of a genuine issue of material fact,” Celotex Corp. v. Catrett, 477 xU.S. 317, 323, 106 S.Ct. 2548 , 91 L.Ed.2d 265 (1986). If the party opposing summary judgment bears the burden of proof at trial, the moving party need only illustrate the “absence of evidence to support the non-moving party’s case.” In re Oracle Corp. Sec. Litig., 627 F.3d 376, 387 (9th Cir.2010). The burden then shifts to the non-moving party to “go beyond the pleadings” and “designate specific facts” in the record to show a trial is necessary to resolve genuine disputes of material fact. Celotex, 477 U.S, at 324, 106 S.Ct. 2548 (quotation marks omitted). “Only disputes over facts that might affect the outcome of the suit under the governing law will properly preclude the entry of summary judgment.” Anderson, 477 U.S. at 248 , 106 S.Ct. 2505 .

The same standard applies to motions for partial summary judgment. See Fed. R. Civ. P. 56(a) (“A party may move for summary judgment, identifying each claim or defense — or the part of each claim or de *955 fense — on which summary judgment is sought — ”); accord Barsamian v. City of Kingsburg, 597 F.Supp.2d 1054, 1061 (E.D.Cal.2009).

The court turns first to the definition of “Governmental Authority” and then considers the definition of “Known Pollution Conditions.”

III. GOVERNMENTAL AUTHORITY

California contract law applies to interpret the policies here. See Bell Lavalin, Inc. v. Simcoe & Erie Gen. Ins. Co., 61 F.3d 742, 745 (9th Cir.1995). Contract interpretation is ordinarily a question of law. Waller v. Truck Ins. Exch., Inc., 11 Cal.4th 1, 18 , 44 Cal.Rptr.2d 370 , 900 P.2d 619 (1995); WYDA Assocs. v. Merner, 42 Cal.App.4th 1702, 1710 , 50 Cal.Rptr.2d 323 (1996). “A contract must be so interpreted as to give effect to the mutuál intention of the parties as it existed at the time of contracting, so far as the same is' ascertainable and lawful.” Cal. Civ. Code § 1636 . A court looks first and foremost to the contract’s language, which controls if “clear and explicit,” provided it “does not involve an absurdity.” Id. § 1638. “When a contract is reduced to writing, the intention of the parties is to be ascertained from the writing alone, if possible....” Id. § 1639.

The ordinary rules of contract interpretation apply just as well to insurance policies. Bank of the W. v. Superior Court, 2 Cal.4th 1254, 1264, 10 Cal.Rptr.2d 538 , 833 P.2d 545 (1992). Unless the parties intended to use words in a technical or special sense, a court reads a contract’s language to understand its plain meaning as a layperson ordinarily would. Waller, 11 Cal.4th at 18 , 44 Cal.Rptr.2d 370 , 900 P.2d 619 (citing Cal Civ. Code § 1638). The policy must also “be read as a whole.” Hartford Accident & Indem. Co. v. Sequoia Ins. Co., 211 Cal.App.3d 1285, 1298 , 260 Cal.Rptr. 190 (1989); see also Cal. Civ. Code § 1641 (“The whole of a contract is to be taken together, so as to give effect to every part, if reasonably practicable, each clause helping to interpret the other.”). Contract interpretations -that deprive words of meaning are to be avoided. See United States v. 1.377 Acres of Land, 352 F.3d 1259, 1265 (9th Cir.2003) (citing Appalachian Ins. Co. v. McDonnell Douglas Corp., 214 Cal.App.3d 1, 12 , 262 Cal.Rptr. 716 (1989)).

Here, LMI requests summary judgment that “the Governmental Authority element of the ELI Policy is satisfied as to all contamination LMI discovers.” Auth. Mem. at 2. “Governmental Authority” is defined in relevant part as follows:

Áuth. Evid. Ex. 1, at SJCE 167594 (bold typeface omitted 6 ). This language clearly explains the parties’ intent: the DTSC agreement is an order and “Government Authority.” Both LMI and Steadfast seem to agree. See Auth. Mem. 5; Auth. Opp’n at 1. The parties also agree op the document that is the DTSC agreement. See Auth. UMF nos. 8-15; Auth. Evid. Ex. 5, ECF *956 No. 160-3. Therefore, LMI’s first motion for summary judgment is granted to the extent it seeks an order that the DTSC agreement is Governmental Authority.

LMI has not carried its burden, however, to show “the Governmental Authority element of the ELI Policy is satisfied as to all contamination LMI discovers.” Auth. Mem. at 2. The DTSC agreement includes twenty-four substantive pages of single-spaced provisions divided into five parts: an introduction; findings of fact (describing Mare Island’s history and the nature and extent of the contamination); a description of the “Response Actions Process” (essentially the tasks to be completed); general provisions (for example on notice and sev-erability); and penalties for noncompliance. See generally Auth. Evid. Ex. 5. It defines its purposes as follows:

Auth. Evid. Ex. 5, ¶ 1.3.

In the DTSC agreement, LMI agreed to conduct certain “response activities” in eight “Investigation Areas.” See id. ¶¶ 3.3-3.16. For example, the DTSC agreement obligates LMI to

It is undisputed these provisions apply to LMI’s clean-up efforts. See Auth. Mem. at 5-7; Auth. Opp’n at 7-8; Auth. Reply at 1. But LMI has not shown it is undisputed its remedial efforts were proposed, reviewed, modified, approved, and executed in line with the DTSC agreement. To the extent this was its goal, the motion is denied. See Auth. Mem. at 8 (“[T]he Governmental Authority element of the ELI Policy coverage grant is satisfied as to any contamination LMI discovers-”). LMI has not satisfied its initial burden to “in *957 form[ ] the district court of the basis for its motion,” Celotex, 477 U.S. at 323 , 106 S.Ct. 2548 .

IV. KNOWN POLLUTION CONDITIONS

A. The Parties’ Positions

LMI’s second motion presents a more complicated question. A more detailed statement of the parties’ positions is necessary as a starting point. LMI requests summary judgment that Known Pollution Conditions “are the conditions specifically identified in the Scope of Work Endorsement to the RSL Policy.” Poll. Mem. at 8. It advocates the following interpretation:

Id. at 7 (emphasis and record citations omitted). LMI acknowledges these general statements are subject to “the [ELI] policy’s definitions, conditions and exclusions.” Id. at 2 n.2.

Steadfast agrees in its opposition brief that the tables and figures listed in the Scope of Work Endorsement “are the beginning of the inquiry,” Poll. Opp’n at 3, and it agrees the definition of Known Pollution Conditions is unambiguous, see id. at 8-9 & n.6. But it believes the list of pollution conditions in those tables is “not the sole arbiter” of whether a .condition is a Known Pollution Condition. Id. at 1. It argues the ELI policy makes several exceptions within its definition of Known Pollution Conditions, and these exceptions show documents other than the Scope of Work Endorsement may influence whether a particular condition is “known.” See id. at 7-10. To show this interpretation is correct, Steadfast cites evidence that LMI, CCI, and Steadfast have referred to “environmental studies” and other documents that were not part of the Scope of Work Endorsement. See id. at 11-18.

Steadfast also argues a pollution condition may be a Known Pollution Condition if it was actually known before the ELI policy took effect, regardless of whether the condition appears in the Scope of Work Endorsement. See id. at 11 (arguing the parties understood they could “resort to environmental studies during the underwriting and during the claims adjusting process ... to determine whether a pollution condition was ’known’ or ’unknown’ ”); id. at 3 (“[T]he issues in this case are not simply whether a condition is listed in a table, but rather whether the actual pollution ... was known.”). Steadfast does not cite any particular policy provision in support of this position; rather, it cites the parties’ correspondence after the policy took effect.

In reply to Steadfast’s opposition brief; LMI reiterates the broad interpretation it espoused in its original motion. See Poll. Reply at 1 (“To be a Known Pollution Condition a condition must appear in the Scope of Work — ”). It also fleshes out a more focused request: that the court find Steadfast cannot “dispute coverage by ... arguing that ’the actual pollution — including, most importantly, its source — was known,’ ” id. at 2 (quoting Poll. Opp’n at 6). In other words, LMI seeks an order clarifying that actual knowledge of pollution on Mare Island does not control whether a pollution condition is “known,” but rather that the ELI policy and the Scope of Work Endorsement alone control that decision. The parties’ supplemental briefing confirms this understanding of LMI’s motion. See, e.g., Suppl. Opp’n at 4 (“The Tables and Figures to the Scope of Work Endorsement reflect the efforts of *958 the parties to create a partial list of what they knew at policy inception to be Known Pollution Conditions.”); Suppl. Reply at 6 (arguing against the position that Steadfast may not “look in every other document” to discover actual knowledge). This dispute has been at the heart of this case for some time. See, e.g., Order May 15, 2015, at 8, ECF No. Ill (“According to LMI, Steadfast seeks to rewrite the policy by arguing there might be conditions that did not make the list of ’knowns’ but which now qualify as known.”).

LMI’s and Steadfast’s conflicting interpretations are unsurprising in light of the incentives they face. Because the ELI policy provides no coverage for Known Pollution Conditions, Steadfast may reduce its liability by adding to the list of Known Pollution Conditions. LMI, oh the other hand, can increase coverage by limiting the number of Known Pollution Conditions.

In resolving LMI’s motion, the court first briefly reviews the effects of its previous order. Second, the court provides a detailed description of the ELI policy’s terms to explain its decision here. Third, the court reviews the substantive law applicable to LMI’s motion, and fourth, it applies that law to the definition of Known Pollution Conditions.

B. The Effect of the Court’s Previous Order'

At hearing Steadfast suggested the court’s previous orders establish the law of the case and require the denial of LMI’s current motion because the court has already rejected LMI’s proposed definition of “Known Pollution Conditions.” 9 This argument is unpersuasive. The court’s previous order expressly invited LMI's current motion. See Order May 15, 2014, at 8, ECF No. Ill (“According to LMI, Steadfast seeks to rewrite the policy by arguing there might be conditions that did not make the list of ’knowns’ but which now qualify as known.,. As the court denied the motion for summary judgment without prejudice, LMI will have a chance to argue this in any renewed motion.”).

C. Summary of Policy Terms

The ELI policy’s definition of “Known Pollution Conditions” includes more than I,000 words. See Evid. Supp. Mot. Summ. J.. Known Pollution Conditions (Poll. Evid), Ex. 2, at SJCP 002583-84, 10 ECF No. 192-2. The definition is inelegant and difficult to parse, but a complete reproduction of its text is necessary to explain the conclusions the court reaches in this order.

1. Foundational Definition

In its first two sentences, the policy lays a foundation:

The definition’s third and fourth sentences provide a more detailed, two-part clarification. First,

In other words, entries in the tables and figures always include “a designation of location, contamination, byproducts, [and] breakdown products.” Some also include a “source of such identified contamination,” and the others include “an expressly urn identified source.”

The definition of Known Pollution Conditions continues:

The definition’s structure suggests the parties’ basic intent: the tables and figures are the center of any inquiry into whether a pollution condition is a- Known Pollution Condition. The policy cites those tables and figures in strong terms: “Known Pollution Conditions are ... those conditions specifically set forth in [the tables and figures]”; “Known Pollution Conditions constitute all of the conditions that are deemed known to the - Insureds for the purposes of this Policy.” But as explained in‘the next paragraphs, the definition allows for exceptions to this general rule.

2. Quantity and Geographic Extent

After laying this groundwork, the definition continues:

This expansion squares with the definition’s first sentences, which specify the tables and figures include “a designation of location [and] contamination....” Notably, if Known Pollution Conditions- include “the entire quantity and geographic extent” of a Known Pollution Condition, then the tables and figures in this sense do not exhaustively define Known Pollution Conditions.

The definition continues by clarifying that the exception described immediately above applies

These provisions anticipate and defuse counterarguments against expansion of the definition of Known Pollution Conditions. In other words, a Known Pollution Condition may expand to encompass its- “entire quantity and geographic extent,” “without regard to,” for example, the. fact a contaminant “migrated across or through one or more environmental media” or. “otherwise had a fate and transport or spatial extent” different from that described in the tables and figures. But because this structure is negative — the exception applies “without regard to” certain facts, essentially depriving those facts of relevance — a party could not affirmatively cite migration across “environmental media” or an unanticipated “fate and transport or spatial extent” to argue a newly discovered pollution condition must necessarily be a Known Pollution Condition. Rather, if a contaminant is later determined to have been a part of a listed Known Pollution Condition’s entire extent, migration across environmental media, or an unexpected spatial extent are irrelevant.

3. Additional Sources and Man-Made Objects

The policy makes two more exceptions. First:

This exception excludes certain costs from the definition of Known Pollution Conditions. In the context of the ELI policy, which covers costs incurred remediating only unknown pollution conditions, the exception expands coverage. But the exception has its own exceptions:

*961 Put differently, “a material misunderstanding” that caused the parties to incorrectly identify “the actual source” of pollution in the tables and figures can prevent the expansion of coverage under this exception. Similarly, if the Scope of Work endorsement originally defined a contaminant’s source as unidentified, but later the source was identified, coverage cannot be expanded under this exception.

Second, the definition includes an exception for interactions with an “unknown man-made subsurface structure or object”:

The policy defines “unknown man-made subsurface structure or object” in the negative:

With this, the ELI policy’s definition of “Known Pollution Conditions” concludes.

4. The Scope of Work Endorsement

As described above, the policy defines Known Pollution Conditions by referring to the Scope of Work Endorsement. The Scope of Work Endorsement provides, “The conditions and activities identified in Tables 1, 2 and 3 and listed below represent the Scope of Work of the Insured Project and are Known Pollution Conditions or actions with respect to such Known Pollution Conditions authorized under the Scope of Work_” Poll. Evid. Ex. 4, at SJCP 029073. It cautions, “The figures illustrate the general locations of contaminants for each site, but do not rep *962 resent the total extent or maximum concentrations, of such contaminant,” Id.

In addition to the tables and figures attached, the endorsement lists several “sections,” which, it provides, “are part of and incorporated into Tables 1 through 3 of this Endorsement, and the contamination conditions and remedial activities identified in these sections shall constitute Known Pollution Conditions for the purposes of this Policy and the Environmental Liability Policy....” Id. at SJCP 029074-75. Section six is titled “Planned Remedial Actions for PCB Sites.” Id. at SJCP 029076, It provides,

D. California Rules Regarding Extrinsic Evidence in Contract Cases

•The' court now turns to the law applicable to the interpretation of the ELI policy. As noted above, California law governs the policy’s interpretation. See Bell Lavalin, 61 F.3d at 745 .

Steadfast rests much of its argument on evidence external to the contract. California law imposes specific limits on evidence other than a contract’s written terms. The court first reviews these rules.

1. The Parol Evidence Rule

Normally a contract’s written terms alone control its interpretation: “when parties enter an integrated written agreement, extrinsic evidence may not be relied upon to alter or add to the terms of the writing.” Riverisland Cold Storage, Inc. v. Fresno-Madera Prod. Credit Ass’n, 55 Cal.4th 1169, 1174 , 151 Cal.Rptr.3d 93 , 291 P.3d 316 (2013) (citing Cal. Code Civ. Proc. § 1856 and Cal. Civ. Code § 1625 ). This rule, commonly called the parol evidence rule, is a rule of substantive law, not of procedure. Casa Herrera, Inc. v. Beydoun, 32 Cal.4th 336, 343 , 9 Cal.Rptr.3d 97 , 83 P.3d 497 (2004). “It is founded on the principle that when the parties put all the terms of their agreement in writing, the writing itself becomes the agreement.” Riverisland, 55 Cal.4th at 1174 , 151 Cal.Rptr.3d 93 , 291 P.3d 316 . Because a contract’s “written terms supersede statements made during the negotiations,” evidence other than those written terms is “irrelevant, and cannot be relied upon.” Id. (emphasis omitted).

The parol evidence rule applies only to “an'integrated written agreement.” Id, A written agreement is “integrated” or is an “integration” if it is “a complete and final embodiment of the terms of an agreement,” Masterson v. Sine, 68 Cal.2d 222, 225 , 65 Cal.Rptr. 545 , 436 P.2d 561 (1968); Ailing v. Universal Mfg. Corp., 5 Cal.App.4th 1412, 1434 , 7 Cal.Rptr.2d 718 (1992), or in other words, if it is “intended by the parties as a final expression of their agreement,” Cal. Code Civ. Proc. § 1856(a). Whether a written contract is an integration is a question of law. Id, § 1856(d); Ailing, 5 Cal.App.4th at 1434 , 7 Cal.Rptr.2d 718 . A court considers several factors when determining whether an agreement is an integration: (1) the *963 presence of an integration clause; (2) the contract’s -language and apparent completeness or incompleteness; (3) if a party argues another contract exists, whether that agreement’s terms contradict those of the written contract; (4) whether the alleged additional agreement would naturally be made as a separate agreement; and (5) whether extrinsic evidence might confuse the jury. Sicor Ltd. v. Cetus Corp., 51 F.3d 848, 859 (9th Cir.1995) (citing McLain v. Great Am. Ins. Cos., 208 Cal.App.3d 1476, 1484 , 256 Cal.Rptr. 863 (1989)); Alling, 5 Cal.App.4th at 1434 , 7 Cal.Rptr.2d 718 .

Here, with respect to the definition of Known Pollution Conditions, the ELI policy is an integrated writing. It is comprehensive, refers expressly to other policies when necessary, and includes the following provision:

Miller Decl. Ex. 1, at 25-26, ECF No. 172-1 (bold typeface omitted). At hearing, the parties confirmed they agree the ELI policy is an integrated agreement, but Steadfast argues extrinsic evidence is nonetheless admissible under exceptions to the parol evidence rule. The court next considers those exceptions

2, Extrinsic Evidence to Establish

and Resolve Ambiguity

Contract law allows admission of extrinsic evidence “to resolve an ambiguity,” even when the contract is an integrated agreement. WYDA, 42 Cal.App.4th at 1710 , 50 Cal.Rptr.2d 323 ; see also Cal. Code Civ. Proc § 1856(g); Winet v. Price, 4 Cal.App.4th 1159, 1165 , 6 Cal.Rptr.2d 554 (1992). Extrinsic evidence may be offered both to explain an obviously ambiguous term and to reveal a latent ambiguity. Pac. Gas & E. Co. v. G.W. Thomas Drayage etc. Co., 69 Cal.2d 33, 37 , 69 Cal.Rptr. 561 , 442 P.2d 641 (1968); Emp’rs Reinsurance Co. v. Superior Court, 161 Cal.App.4th 906, 920 , 74 Cal.Rptr.3d 733 (2008).

“An agreement is not ambiguous merely because the parties (or judges) disagree about its meaning.” Abers v. Rounsavell, 189 Cal.App.4th 348, 356 , 116 Cal.Rptr.3d 860 (2010). Rather, the determination of ambiguity involves two steps. “First, the court provisionally receives (without actually admitting) all credible evidence concerning the parties’ intentions to determine ’ambiguity,’ i.e., whether the language is ’reasonably susceptible’ to the interpretation urged by a party.” Winet, 4 Cal.App.4th at 1165 , 6 Cal.Rptr.2d 554 . If the contract is not “ ’reasonably susceptible’ to the interpretation urged,” then “the case is over.” S. Cal. Edison Co. v. Superior Court, 37 Cal.App.4th 839, 847-48 , 44 Cal.Rptr.2d 227 (1995). But “[i]f in light of the extrinsic evidence the court decides the language is ’reasonably susceptible’ to the interpretation urged, the extrinsic evidence is then admitted to aid in the second *964 step — interpreting the contract.” Winet, 4 Cal.App.4th at 1165 , 6 Cal.Rptr.2d 554 .

Because extrinsic evidence must be “relevant to prove a meaning to which the language of the contract is reasonably susceptible,” any extrinsic evidence must be tethered to specific contract language. Alameda Cnty. Flood Control v. Dep’t of Water Res., 213 Cal.App.4th 1163, 1188-89 , 152 Cal.Rptr.3d 845 (2013) (quoting Dore v. Arnold Worldwide, Inc., 39 Cal.4th 384, 391 , 46 Cal.Rptr.3d 668 , 139 P.3d 56 (2006)) (emphasis omitted). Whether a contract is ambiguous is a question of law, but if the contract is ambiguous, the conflict is ordinarily a genuine dispute of material fact inappropriate for resolution on summary judgment. San Diego Gas & Elec. Co. v. Canadian Hunter Mktg. Ltd., 132 F.3d 1303 , 1307 (9th Cir.1997); WYDA, 42 Cal.App.4th at 1710 , 50 Cal.Rptr.2d 323 .

3. Evidence of a Course of Performance

One form of extrinsic evidence is the contracting parties’ “course of performance.” 12 See Cal. Civ. Proc. Code § 1856 (c); Epic Commc’ns, Inc. v. Richwave Tech., Inc., 237 Cal.App.4th 1342, 1355 , 188 Cal.Rptr.3d 844 (2015); Emp’rs Reinsurance, 161 Cal.App.4th at 920 , 74 Cal.Rptr.3d 733 . Section 1303 of the California Commercial Code defines the term “course of performance” for purposes of an insurance policy dispute like this one. See Cal. Civ. Proc. Code § 1856 (c) cmt. to 1978 am.; Emp’rs Reinsurance, 161 Cal.App.4th at 920 , 74 Cal.Rptr.3d 733 (noting a renumbering of the relevant code sections). “A ’course of performance’ is a sequence of conduct between the parties to a particular transaction .... ” Cal. Com. Code § 1303 (a). It exists if both (1) the parties’ agreement “involves repeated occasions for performance” by one of them and (2) the other has “knowledge of the nature of the performance and opportunity for objection to it” but “accepts the performance or acquiesces in it without objection.” Id.

“The rationale for the admission of course of performance evidence is a practical one.” Emp’rs Reinsurance, 161 Cal.App.4th at 921 , 74 Cal.Rptr.3d 733 . When interpreting a contract, the court’s duty is to give effect to the parties’ intentions at the time they entered the contract. Cal. Civ. Code § 1636 . And “the most reliable evidence of the parties’ intentions” is their behavior after the contract is signed and before any controversy has arisen. Kennecott Corp. v. Union Oil Co. 196 Cal.App.3d 1179, 1189 , 242 Cal.Rptr. 403 (1987).

In federal court, “[a]n inference of the parties’ common knowledge or understanding that is based upon a prior course of dealing is a question of fact.” In re CFLC, Inc., 166 F.3d 1012, 1017 (9th Cir.1999). In light of the Ninth Circuit’s reasoning in In re CFLC, the same may likely be said of the parties’ course of performance. See id. (citing New Moon Shipping Co. v. MAN B & W Diesel AG, 121 F.3d 24, 31 (2d Cir.1997) (discussing the factfinder’s role in drawing inferences from the contracting parties’ actions)). A genuine dispute surrounding the parties’ course of performance may therefore preclude summary judgment.

4. The Role of Ambiguity

Recent decisions in California appellate courts contradict one another on the role of ambiguity in the admissibility of course of performance evidence. In Employers *965 Reinsurance Co. v. Superior Court, the California Court of Appeal explained the parties’ course of performance will be adopted and enforced only when it is a reasonable interpretation of the contract’s terms. 161 Cal.App.4th at 921 , 74 Cal.Rptr.3d 733 . Later the court repeated this rule, explaining a course of performance could be admitted “regardless of the actual language of the contract, as long as the parties’ interpretation is reasonable.” Id. at 922 , 74 Cal.Rptr.3d 733 (emphasis omitted). In the same paragraph, however, the court summarized that “course of performance evidence can supplement, qualify, or modify contrary terms in the contract.” Id. This holding seems to refer to California Commercial Code section 1303(f), which the state appellate court had quoted earlier in its decision: evidence of a course of performance is relevant to “show a waiver or modification of any term inconsistent with the course of performance.” Id. at 920-21 , 74 Cal.Rptr.3d 733 (quoting Cal. Com. Code § 1303 (f)).

It is difficult to understand how a course of performance could be both a reasonable interpretation of a contract and “inconsistent with” the contract’s written terms. These citations are also difficult to square with the same court’s description of course of performance evidence as a type of extrinsic evidence, which it emphasized “is admissible when relevant to prove a meaning to which the language of the instrument is reasonably susceptible.” Id. at 920 , 74 Cal.Rptr.3d 733 (citation and quotation marks omitted). In a later decision, a different panel of the same Court of Appeal suggested the Employers court’s citation to section 1303(f) was dicta, only “a passing reference.” Lonely Maiden Prods., LLC v. GoldenTree Asset Mgmt., LP, 201 Cal.App.4th 368, 378 , 135 Cal.Rptr.3d 69 (2011).

In 2010, another Court of Appeal quoted Employers and confirmed course of performance evidence, like any extrinsic evidence, “is admissible when relevant to prove a meaning to which the language of the instrument is reasonably susceptible.” In re Tobacco Cases I, 186 Cal.App.4th 42, 52 , 111 Cal.Rptr.3d 313 (2010) (quoting 161 Cal.App.4th at 920 , 74 Cal.Rptr.3d 733 ). In Tobacco Cases I , the court rejected evidence of the parties’ course of performance offered to explain the word “cartoon” because the agreement’s written terms were not susceptible to the interpretation the course of performance would have supported. Id.

Like Employers, however, Tobacco Cases I also has come under criticism in later opinions, albeit in dicta. In Epic Communications, Inc. v. Richwave Technology, Inc., the court concluded a settlement agreement’s release clause was “patently ambiguous.” 237 Cal.App.4th at 1354 , 188 Cal.Rptr.3d 844 . Therefore the court found it proper to consider the parties’ course of performance as evidence of their intent. Id. at 1354-55 , 188 Cal.Rptr.3d 844 . Although the court had already concluded the contract terms in question were patently ambiguous, it quoted the Law Revision Committee’s comments to section 1856(c): “Subdivision (c) [of ' California Code of Civil Procedure § 1856] definitely rejects... the requirement that a condition precedent to the admissibility of the type of evidence specified in the subdivision is an original determination that the language used is ambiguous.” Id. at 1355, 188 Cal.Rptr.3d 844 (citing Cal. Civ. Proc. Code § 1856 , cmts. to 1978 am.) (ellipses in Epic). In a footnote, the Epic court then rejected the “more stringent rule” of Tobacco Cases I , which “appealed] to rest on the premise that course-of-performance evidence is admissible only if the contract is first found to be ambiguous, or at least latently ambiguous.” Id. at 1355, 188 Cal.Rptr.3d 844 n. 13 .

*966 In the presence of these competing authorities, this court must anticipate how the California Supreme Court would resolve the issue. See, e.g., Kairy v. Super-Shuttle Int’l, 660 F.3d 1146, 1150 (9th Cir.2011). California Code of Civil Procedure section 1856 is the common thread in the cases cited above. The court therefore looks first to that section. Subdivision ■ (a) provides, “Terms set forth in a writing intended by the parties as a final expression of their agreement with respect to the terms included therein may not be contradicted by evidence of a prior agreement of or a contemporaneous oral agreement.” Cal. Civ. Proc. Code § 1856 (a). Subdivision (c) allows “[t]he terms set forth in a writing described in subdivision (a) [to] be explained or supplemented ... by course of performance,” Id. § 1856(c).

The words “explained” and “supplemented” are not synonyms of “contradicted.” Moreover, California Commercial Code section 2202, which the Law Revision Commission Comments describe as “comparable to” Code of Civil Procedure section 1856, also uses the words “explained. or supplemented” to describe permissible uses of course of performance evidence and “contradicted” to. describe impermissible uses of extrinsic evidence. Cal. Com. Code § 2202 . See also Apex LLC v. Sharing World, Inc., 206 Cal.App.4th 999, 1014 , 142 Cal.Rptr.3d 210 (2012) (“[Under section 2202,] to the extent the writing incorporates terms the parties agreed on, then the writing controls as to those terms and anything contradictory or inconsistent is of no force or effect.” (citation and quotation marks omitted)). This is also the general rule of the Uniform Commercial Code. See 2A Anderson U.C.C. § 2-208:5 (3d. ed. updated 2015) (“Evidence of a course of performance is admissible if it does not directly contradict the terms of a written agreement, but merely explains or supplements them.”). And courts outside California also regularly interpret analogous state law similarly. See, e.g., Ellicott Mach. Corp. Int’l v. Jesco Const. Corp., 199 F.Supp.2d 290, 295 (D.Md.2002); Craig Sessions, M.D., P.A. v. TH Healthcare, Ltd., 412 S.W.3d 738, 745-46 (Tex.App.2013); Noble Roman’s, Inc. v. Pizza Boxes, Inc., 835 N.E.2d 1094, 1099 (Ind.Ct.App.2005); A & A Mech., Inc. v. Thermal Equip. Sales, Inc., 998 S.W.2d 505, 510 (Ky.Ct.App.1999); J.R. Waymire Co. v. Antares Corp., 975 S.W.2d 243, 247 (Mo.Ct.App.1998); KN Energy, Inc. v. Great W. Sugar Co., 698 P.2d 769, 779 (Colo.1985).

This court concludes section 1856(c) makes course of performance evidence admissible to explain or supplement but not to contradict the terms of an integrated agreement, even when the agreement’s written terms are unambiguous. The distinction between evidence that explains and evidence that contradicts is familiar in California jurisprudence. See, e.g., Hervey v. Mercury Cas. Co., 185 Cal.App.4th 954, 961 , 110 Cal.Rptr.3d 890 (2010) (“Although parol evidence may be admissible to determine whether the terms of - a contract are ambiguous, it is not admissible if it contradicts a clear and explicit policy provision.” (citation omitted)); Winet, 4 Cal.App.4th at 1167 , 6 Cal.Rptr.2d 554 (“[Pjarol evidence is admissible only to prove a meaning to which the language is ’reasonably susceptible,’ not to flatly contradict the express terms of the agreement.” (citation omitted)). Moreover, section 1856(c) “codifies prior law.” Cal. Civ. Proc. Code § 1856 cmt. to 1978 am. California courts have long held parol evidence may not contradict the terms of an integrated agreement. See, e.g., Masterson, 68 Cal.2d at 225-26 , 65 Cal.Rptr. 545 , 436 P.2d 561 (parol evidence cannot “vary” the terms of a written agreement); Universal Sales Corp. v. California Press Mfg. Co., *967 20 Cal.2d 751 , 128 P.2d 665 (1942) (evidence of the parties’ “practical construction” of a contract is not admissible when it would only support an interpretation “wholly at variance with the correct legal interpretation” of the contract’s terms).

This reading of section 1856(c) is consistent with Epic Communications, Employers, and Tobacco Cases I . According to section 1856’s comments, an ambiguity is not a prerequisite to admission under section 1856(c) if the evidence will explain and supplement . a contract. See Epic Commc’ns, 237 Cal.App.4th at 1355 n. 13, 188 Cal.Rptr.3d 844 . And according to the Employers and Tobacco Cases I courts, parol evidence may not flatly contradict the written contract or support an unreasonable interpretation of it. See 186 Cal.App.4th at 52 , 111 Cal.Rptr.3d 313 (quoting 161 Cal.App.4th at 920 , 74 Cal.Rptr.3d 733 ).

Here, as a practical matter, the court concludes evidence of the parties’ course of performance is admissible even if Steadfast makes no effort to show the contract is ambiguous and even if the court makes no Ending'to that effect. The Law Review Commission’s meticulous phrasing suggests this interpretation. See Cal. Civ. Proc. Code § 1856 cmt'. to 1987 am. (rejecting “the requirement that a condition precedent to the admissibility of [course of performance evidence] is an original determination that the language used is ambiguous.”). Nevertheless, LMI may challenge the evidence by arguing it could support no reasonable interpretation of the ELI Policy or flatly contradicts the policy. If evidence does flatly contradict the policy, the question would presumably be whether the contract was modified, an issue the parties have not raised or briefed, and one the court does not consider here. See, e.g., Cal. Civ. Code § 1698 ; Wagner v. Glendale Adventist Med. Ctr., 216 Cal.App.3d 1379, 1388 , 265 Cal.Rptr. 412 (1989).

E. Discussion

The law described above leaves the court with three questions: (1) whether the ELI Policy’s written terms support the interpretation LMI proposes; (2) whether Steadfast’s extrinsic evidence is admissible to resolve an ambiguity in the policy; and (3) whether Steadfast’s extrinsic evidence is admissible to establish a course of performance consistent with Steadfast’s interpretation of the policy.

1. The ELI Policy’s Written Terms

To decide whether a condition is a Known Pollution Condition, the policy refers . twice to the conditions “specifically described” or “specifically set forth” in the Scope of Work Endorsement. Not only does the. policy, refer repeatedly to the Scope of Work endorsement, it often repeats the lengthy reference to “Tables 1-3 and Figures 1 through 89 and Fuel Oil Line Removal Project Figures 1 through 11 to the Scope of Work Endorsement.” Even when the policy makes exceptions it refers to the same tables and figures. This choice of language suggests the parties intended to cabin any discussion of known and unknown conditions to the Scope of Work endorsement.

But the policy is- more complex. The definition of Known Pollution Conditions includes the “entire quantity and . geographic extent” of a contaminant later determined to have been part of a Known Pollution Condition — regardless of “whether the contaminant ... was identified in [the tables and figures] in quantities or concentration(s) requiring remedial action at the time it was identified” and regardless of whether it “had a fate and transport or spatial extent different than that understood as set forth in” the tables and figures. In this light, the Scope. of Work endorsement and the tables and figures do not exhaustively define Known Pollution Conditions. Otherwise it would be impossi *968 ble to “ultimately determine[ ]” a contaminant, although not originally listed, was actually part of the “entire quantity and geographic extent” of a Known Pollution Condition and therefore “deemed” a Known Pollution Condition.

Similarly, the exceptions for additional sources and any “unknown man-made subsurface structure or object” reach beyond the Scope of Work Endorsement and attached tables and figures. Otherwise it would be impossible to determine (1) whether an “incorrect identification indicates a material misunderstanding by the Company and an Insured as to the actual source of the contaminants” listed in the tables and figures, or (2) if “an increase in extent, concentration or quantity of a contaminant over' that identified in the” tables and figures was “the direct result of contaminant interaction with an ’unknovm man-made subsurface structure or object.’ ”

The court also notes the parties’ apparent agreement that the tables and figures in the Scope of Work endorsement “are often meaningless without reference to the source documents” and that the parties may need to “resort ... to contemporaneous technical documents” and “historic environmental documents ... to determine what was meant by reference to a given ’location,’ ’contaminant’ and ’source.’” Poll. Opp’n at 3; Poll. Reply at 1 (“[D]ocu-ments external to the policy may be relevant to explain an unclear entry on the tables ....”). After all, evidence extrinsic to the contract is generally admissible to explain its meaning if the written terms are reasonably susceptible to that meaning. Casa Herrera, 32 Cal.4th at 343 , 9 Cal.Rptr.3d 97 , 83 P.3d 497 .

In sum, the court cannot agree with LMI that to determine whether a pollution condition is known or unknown, one need only “look on the Scope of Work” and that “[n]o further inquiry is appropriate.” Poll. Mem. at 7. The policy’s definition is not so simple.

On the other hand, the ELI Policy’s written definition of Known Pollution Conditions does support LMI’s more focused contention that a pollution condition cannot be “known” simply because Steadfast can demonstrate historical knowledge of it. Were the opposite true, then the following terms would be meaningless:

The painstakingly detailed exceptions that follow these sentences reinforce this conclusion. Had the parties intended actual knowledge to control rather than the Scope of Work Endorsement, a much shorter definition saying as much would be expected. And although Steadfast points out several exceptions within the definition of Known Pollution Conditions, none allows the conclusion that actual knowledge controls. Each exception refers to the Scope of Work Endorsement.

In Steadfast’s supplemental brief, it argues this interpretation would deprive the ELI Policy’s Coverage A provision of meaning. See Supp. Opp’n at 4. As noted above, see supra note 4, the ELI Policy provides coverage only for pollution conditions “first discovered by an Insured during the Policy Period.” Steadfast did not raise this, argument in its opposition brief, despite the court’s previous order express *969 ly reserving consideration of that provision for this motion. See Order May 15, 2014, at 8-9 (“Steadfast argues the omission does not matter because the court said this ’deemed known’ provision would render the ’first discovered’ provision surplus-age _LMI argues the ’first discovered language,’ when read in conjunction with the ’deemed known’ provision, operates to limit the coverage to contamination discovered before the end of the policy period.... Whatever the ultimate meaning of this provision, it is better resolved on a renewed motion for summary judgment.”). By raising this argument only belatedly now, Steadfast signaled the court did not need to raise it with the parties at hearing, and LMI was left only its supplemental reply brief for a rebuttal. See Suppl. Reply at 5 n.3. Problems like this motivate Local Rule 230, which establishes a briefing and hearing schedule for all motions. See E.D. Cal. L.R. 230(a)-(d).

Nevertheless, because the court does not accept Steadfast’s argument, considering it here will not prejudice LMI. In short, Steadfast’s interpretation of Coverage A would render the definition of Known Pollution Conditions superfluous. The parties would have had no reason to construct an intricate, thousand-word definition if discovery of any pollution condition before the policy period automatically excluded coverage. It is also possible to interpret the provisions of Coverage A to avoid a conflict. See Cal. Civ. Code § 1652 (“Re-pugnancy in a contract must be reconciled, if possible, by such an interpretation as will give some effect to the repugnant clauses, subordinate to the general intent and purpose of the whole contract.”); id. § 1653 (‘Words in a contract which are wholly inconsistent with its nature, or with the main intention of the parties, are to be rejected.”). Requiring conditions be discovered “during the Policy Period,” rather than after the Policy Period, is unsurprising in light of the ELI Policy’s extended reporting periods, which the Coverage A provision expressly cites. 13

In summary, first, LMI’s motion is denied to the extent it argues one need only “look on the Scope of Work” and “[n]o further inquiry is appropriate.” Poll. Mem. at 7. The definition of Known Pollution Conditions includes exceptions that may require documentary investigation beyond the Scope of Work Endorsement. Documents external to the policy may also be necessary to shed light on the meaning of an item in the Scope of Work Endorsement or entries in tables and figures. Second, Steadfast cites no policy term showing a pollution condition may become a Known Pollution Condition on the basis of actual knowledge only. The court therefore turns to evidence extrinsic to the ELI Policy, but only with respect to this second question.

2., Extrinsic Evidence

The court first summarizes Steadfast’s proposed extrinsic evidence, then considers whether that evidence is admissible to resolve an ambiguity or to establish the parties’ course of performance.

a) Proposed Evidence

Steadfast divides its parol evidence into five categories. First, it offers excerpts from the deposition testimony of Karen Lubovinsky, the person Steadfast designated as its representative under Federal Rule of Civil Procedure 30(b)(6). Poll. Opp’n at 11-12. Ms. Lubovinksy testified that in her experience, CCI “looked at *970 historic Navy documents” and “all of [the] historic record becomes part of the evaluation of what [CCI] is doing and why they’re doing it.” Lubovinsky Dep. 207-08, ECF No. 212-1.

Second, Steadfast offers evidence that LMI and CCI have referred to documents other than the Scope of Work endorsement to decide whether a pollution condition was a Known Pollution Condition. Poll. Opp’n at 12-15. For example, Steadfast presents evidence in the form of a letter from LMI to Steadfast that in January 2010, LMI understood a pollution condition was a Known Pollution Condition in part because it “was known to exist at the time of policy inception,” id. at 12, and in May 2009, LMI referred to its “records from the time of policy underwriting,” id. at 15 (emphasis omitted).

Third, Steadfast offers evidence LMI and CCI have disagreed whether the tables and figures in the policy were exhaustive. Id. at 16-17. For example, Steadfast cites evidence that CCI was concerned about its dispute with LMI over “what constitutes an Unknown,” id, at 16, and in past disputes with CCI, LMI had concurred with Steadfast that a pollution condition could be unknown as provided in various “underwriting materials,” id. at 16.

Fourth, Steadfast offers evidence LMI understands the difficulty of determining whether a pollution condition is a Known Pollution Condition. Poll. Opp’n at 17-18. For example, Steadfast cites evidence LMI acknowledged in 2011 that some pollution conditions did not “fall[ ] neatly into known or unknown categories,” and that pollution conditions can be “partially known and partially unknown.” Id. at 18 (emphasis omitted).

Fifth, Steadfast argues section six of the Scope of Work endorsement is ambiguous. Poll. Opp’n at 9-10. As summarized above, the section describes five “planned remedial actions,” including “PCB characterization ... at 84 additional sites” and “PCB abatement ... at 42 additional sites.” Poll. Evid. Ex. 4, at SJCP 029076. The section includes no other description or identification of these 126 “additional sites.” Steadfast relies on evidence “the parties met in 2005 to reach agreement on exactly what those ’known’ sites included, taking into consideration the Site Summary Forms and other historic documents.” Poll. Opp’n at 10.

b) Ambiguity

The court first “provisionally receives (without actually admitting) all credible evidence concerning the parties’ intentions to determine ’ambiguity,’ ie., whether the language is ’reasonably susceptible’ to the interpretation urged by a party.” Winet, 4 Cal.App.4th at 1166 , 6 Cal.Rptr.2d 554 . If the ELI policy is not “reasonably susceptible” to the interpretation Steadfast urges, then the evidence cannot be admitted. S. Cal. Edison Co., 37 Cal.App.4th at 847-48 , 44 Cal.Rptr.2d 227 .

Steadfast must tie its evidence “to particular language” of the written agreement; this is “essential.” Alameda Cnty. Flood Control, 213 Cal.App.4th at 1188 , 152 Cal.Rptr.3d 845 (emphasis omitted); cf., e.g., Dore, 39 Cal.4th at 392 , 46 Cal.Rptr.3d 668 , 139 P.3d 56 (considering whether the phrase “at will” was -ambiguous); Am. Alt. Ins. Corp. v. Superior Court, 135 Cal.App.4th 1239, 1247 , 37 Cal.Rptr.3d 918 (2006) (considering whether the phrases “physical damage” and “damaged property” were ambiguous); Pac. Gas & Elec., 69 Cal.2d at 39 & n. 6, 69 Cal.Rptr. 561 , 442 P.2d 641 (considering whether an indemnity clause was ambiguous and citing cases where the court had considered whether the words “United Kingdom,” “ton,” “stubble,” and “north” were ambiguous); Epic Commc’ns, 237 Cal.App.4th at 1348-54 , 188 Cal.Rptr.3d 844 (considering conflicting sentences); To *971 bacco Cases I, 186 Cal.App.4th. at 50-52, 111 Cal.Rptr.3d 313 (considering whether the word “cartoon” was ambiguous); In re Black’s Estate, 211 Cal.App.2d 75, 85 , 27 Cal.Rptr. 418 (1962) (considering whether the phrase “To The University of Southern California known as The U.C.L.A” was ambiguous).

Apart from the fifth category of parol evidence, Steadfast does not tie its evidence to any specific term, phrase, or clause. Neither does Steadfast address the ELI Policy’s directives that (1) “Known Pollution Conditions means all conditions specifically described in the Scope, of Work Endorsement to the Remediation Stop Loss Policy;” and (2) “Known Pollution Conditions constitute all of the conditions that are deemed known to the Insureds for the purposes of this Policy.” Rather, Steadfast simply argues these phrases cannot mean what they say. See, e.g., Poll. Opp’n at 3 (“[T]he issues in this ease are not simply whether a condition is' listed in a table, but rather whether the actual pollution ... was known.”). But parol evidence is not admissible to “flatly contradict the express terms of the agreement.” Winet, 4 Cal.App.4th at 1167 , 6 Cal.Rptr.2d 554 (citing Stevenson v. Oceanic Bank, 223 Cal.App.3d 306, 317-18 , 272 Cal.Rptr. 757 (1990)). The court therefore cannot consider evidence from the first four categories to “explain or otherwise shed light on the parties’ intent.” Thrifty Payless, Inc. v. Mariners Mile Gateway, LLC, 185 Cal.App.4th 1050, 1061 , 111 Cal.Rptr.3d 173 (2010).

As to the fifth category, Steadfast argues section six of the Scope of Work endorsement is ambiguous because its text refers to 126 “additional sites” without identifying them. Poll. Opp’n at 10. Steadfast relies on a declaration submitted by Ms. Lubovinsky. See Statement Disputed Facts no. 19, ECF No. 194 (citing Lubovinsky Deck ¶ 5, ECF No. 197). Ms. Lubovinsky recounts- a March 14, 2005 meeting with about twenty representatives from CCI, LMI, and Steadfast. Lubovinsky Deck ¶ 5. The purpose of the meeting was to “come to agreement on whether more than one hundred PCB claims involved ’known’ or ’unknown’ contamination.” Id. At the meeting, the parties considered “certain Navy Technical Memoranda and Environmental Documents and other pre-policy documentation.” Id. The. parties decided some of these “PCB claims” were “known,” some were “unknown,” and reached no conclusion as to others. Id. “[B]oth CCI and LMI made clear during that meeting that these documents were critical to the resolution.” Id. CCI sent Ms. Lubovinksy an email before the meeting, which attached “additional information made available by the Navy” about “115 sites so as to enable Zurich to make a determination as to whether they were known or unknown.” Id. Ex. B, at SZC 101531. Therefore, Steadfast reasons, the court must refer to “the historic environmental documents.” Id.

First, Steadfast has not explained how a finder of fact could conclude the 115 sites Ms. Lubovinsky refers to in her declaration are the same sites as the 126 sites mentioned in section six of the Scope of Work Endorsement. Without this explanation, its evidence lacks foundation and cannot support Steadfast’s position.

Second, assuming for sake of argument this numerical discrepancy could be resolved, to show the Scope of Work endorsement is ambiguous, Steadfast must present evidence that the Scope of Work endorsement is reasonably susceptible to multiple interpretations. For example, Steadfast could present evidence -that section six could be interpreted to refer to multiple physical locations. It could show the additional sites may or may not be *972 among those listed in the tables and figures. Or if the additional sites are among those listed in the tables and figures, Steadfast could show it is unclear on what page and in which row. The evidence about the 2005 meeting establishes no ambiguity about section six’s language. Rather, Steadfast presents the March 2005 meeting as evidence that the parties at one time considered actual knowledge sufficient to create a Known Pollution Condition. For the reasons described above, parol evidence of this type is not admissible here.

Third, this evidence sheds no persuasive light on the text in question. In relevant part, section six provides as follows: “The planned remedial actions ... for the sites listed on Table 3 and shown on Figures 84 through 88 are as follows ... 3. Perform PCB characterization as required by Governmental Authority at 84 additional sites .... 4. Perform PCB abatement as required by Governmental Authority of 42 additional sites_” Poll. Evid. Ex. 4, at SJCP 029076. Items three and four are among the “planned remedial actions ... for the sites listed on Table 3 and shown on Figures 84 through 88.” Id. That is, the 84 and 42 “additional sites” are among those listed in the tables and figures, whether or not the parties considered other documentation at a March 2005 meeting, and the text is not ambiguous. See Alameda Cnty. Flood Control, 213 Cal.App.4th at 1189 , 152 Cal.Rptr.3d 845 (“[W]ritten agreements whose language appears clear in the context of the parties’ dispute are not open to claims of ’latent’ ambiguity.” (citations and internal quotation marks omitted)).

Steadfast has not shown any particular policy language is reasonably susceptible to more than one interpretation. Its parol evidence is therefore inadmissible to explain any ambiguity.

c) Course of Performance

As the court has concluded above, section 1856(c) of the California Code of Civil Procedure allows Steadfast to present evidence of the parties’ course of performance even without showing a contract term is ambiguous. As described above, “course of performance,” is defined in particular as follows:

Cal. Com. Code § 1303 (a); accord Epic Commc’ns, 237 Cal.App.4th at 1355 , 188 Cal.Rptr.3d 844 . In other opinions, California courts have expressed more concretely that course of performance evidence “applies only to acts performed under the contract before any dispute has arisen.” Warner Constr. Corp. v. City of L.A., 2 Cal.3d 285, 297 , 85 Cal.Rptr. 444 , 466 P.2d 996 (1970). In other words, actions undertaken after the parties reach “a stage of clear disagreement” are not admissible under this rule. Id.

Steadfast must therefore make four preliminary showings before its evidence is admissible. First, the ELI Policy must contemplate a series of repeated performance by a particular person or entity. Second, that person or entity must be shown to have acted in accordance with Steadfast’s current interpretation of the ELI Policy. Third, Steadfast must show LMI knew about these actions and had an opportunity to object, but did not. And fourth, all this must have occurred after *973 the ELI Policy was executed and before the parties to the ELL Policy reached “a stage of clear disagreement.” Overarching this test is the general rule, that evidence of the parties’ course of performance is not admissible to flatly contradict the written policy. See Universal Sales, 20 Cal.2d at 761-62 , 128 P.2d 665 ; Tobacco Cases I, 186 Cal.App.4th at 52 , 111 Cal.Rptr.3d 313 ; Emp’rs Reinsurance, 161 Cal.App.4th at 921 , 74 Cal.Rptr.3d 733 .

Steadfast has not cited the record to show any genuine disputes of material fact prevent summary judgment in LMI’s favor as to actual knowledge. As a preliminary matter, much of Steadfast’s evidence simply does not speak to a possible course of performance. Several of Steadfast’s citations speak only to CCI’s or Steadfast’s knowledge, actions, or positions, rather than LMI’s. See Miller Decl. Ex. E, at 14, ECF No. 202-5 (describing a dispute between CCI and Steadfast); id. Ex. H, at 1, ECF No. 202-8 (reporting CCI’s position); id. Ex. J, at 2, ECF No. 202-10 (reporting CCI’s position); id. Ex. L, at 1, ECF No. 202-12 (describing CCI’s actions); id. Ex. N, at l 14 (describing CCI’s actions); id. Ex. O, at 1, ECF No. 202-14 (CCI states its understanding); id. Ex. U, at 1-2, ECF No. 208-2 (LMI states its belief Steadfast will likely consider a condition known on the basis of documentation outside the Scope of Work Endorsement); id. Ex. Y, at 3, ECF No, 208-6 (LMI states its understanding of a conflict between CCI and Zurich and founds its own position on the basis of contracts other than the ELI Policy); id. Ex. II, at 2-3, ECF No. 211-1 (CCI categorizes unknown PCB sites); id. Ex. JJ, at 187-88, ECF No. 212-1 (CCI relied on documents other than the Scope of Work Endorsement).

Other citations are simply irrelevant. See id. Ex. W, at 1, ECF No. 208-4 (LMI pillories CCI for relying on arguments based on the parties’ intent rather than contract provisions); id. Ex. X, at 5, ECF No. 208-5 (CCI reports its belief that disputes with LMI “as to what constitutes an Unknown” are a risk management issue); id.- Ex. AA, at 1, ECF No. 209-1 (LMI argues CCI faced conflicts of interest that drove CCI to mischaracterize pollution conditions as unknown).

Several of Steadfast’s citations suggest other documentation may help interpret the Scope of Work Endorsement, but do not suggest actual knowledge controls the definition of Known Pollution Conditions. See id. Ex. F, ECF No. 202-6 (LMI relied on the Scope of Work Endorsement to conclude a condition is unknown and supported this conclusion with other documentation); id. Ex. G, at 1-2, ECF No. 202-7 (LMI relates its understanding of why a condition was included in the Scope of Work endorsement); id. Ex. V, at 1-2, ECF No. 208-3 (LMI states its belief that a condition is known based both on its listing in the Scope of Work Endorsement and historical documentation).

And several of Steadfast’s citations suggest costs may be allocated between Known Pollution Conditions and unknown conditions. See id. Ex. CC, at 1, ECF No. 209-3 (LMI writes that distinction between known and unknown pollution conditions is murky); id. Ex. DD, at 1, ECF No. 209-4 (LMI writes some sites do not fall “neatly into known or unknown categories”); id. Ex. EE, at 1, ECF No. 209-5 (LMI describes “co-located Known and Unknown contamination”); id. Ex. FF, at 1, ECF No. 209-6 (CCI writes “costs must be properly apportioned between the RSL and ELI policies”); id. Ex. GG, at 1, ECF No. 209-7 (LMI writes it is “trying to come up with a way of allocating the costs of known/un *974 known conditions between the RSL/ELI policies”). But this conclusion is consistent with the ELI Policy’s definition, which provides that “Known Pollution Conditions shall not include the allocable portion of costs attributable to completing the Scope of Work with respect to any contaminant ..,, to the extent that the contaminant results from a source that is in addition to a source for that contaminant specified in the Scope of Work endorsement.” Poll. Evid. at SJCP 002584.

Finally, the remaining citations could be relevant to establish LMI’s knowledge that other contract parties valued actual knowledge over a listing in the Scope of Work Endorsement. See Miller Decl. Ex. M, at 3, ECF No. 202-13 (CCI summarizes its understanding to this effect in correspondence to LMI); id. Ex. P, at 2, ECF No. 202-15 (same); id. Ex. Q, at 2, ECF No, 202-16 (same); id. Ex. S, at 2, ECF No. 210-1 (same); id. Ex. BB, at 2, 4, ECF No. 209-2 (same). Alternatively, some evidence suggests LMI acquiesced in that understanding. See id. Ex. E, at 16 (CCI reports its understanding of LMI’s position that “Zurich determines Khowns/Unknowns”); idr Ex. R, at 1, ECF No. 202-17 (“LMI claims Known [Fuel Oil Pipeline or] FOPL segments are those shown on Figures in [another' agreement, the Environmental Services Cooperative Agreement or] ESCA ... CH2M HILL claims only those segments shown on Table 2 of the RSL Policy define the -list of Known FOPL segments. LMI using ESCA to define Known FOPL segments ... CH2M HILL using the-RSL Policy to define Known FOPL segments.”); Lubovinksy Decl. ¶ 6, ECF No. 197 (reporting “LMI has often agreed that resolution of whether given pollution was ’known’ or ’unknown’ necessarily turns upon what the parties knew prior to the policies’ inception, regardless of whether a given pollution condition is expressly listed in the tables appended to the policies”).

But fundamentally, Steadfast has cited no portion of the record to show its evidence depicts relevant performance before any controversy arose. It makes ho attempt to delineate a time period in which the parties “harmoniously performed the contract.” Emp’rs Reinsurance, 161 Cal.App.4th at 922 , 74 Cal.Rptr.3d 733 . Rather, Steadfast’s citations are replete with references to disputes about the meaning of Known Pollution Conditions. See, e.g., Miller Decl. Ex. U, at 1-2; id. Ex. W, at 1; id. Ex. X, at 5; id. Ex. Y, at 3; id. Ex. AA, at 1; Lubovinsky Decl. ¶ 5 (describing 2005 meeting meant to resolve disagreements about definition of Known Pollution Conditions). Its course of performance evidence is therefore inadmissible. 15

Furthermore, as the court concluded above, Steadfast’s fundamental position flatly contradicts the terms of the ELI Policy. Steadfast argues “the issues in this case are not simply whether a condition is listed in a table, but rather whether the actual pollution ... was known.” Poll. Opp’n at 3. It argues the Scope of Work Endorsement is only “a partial list of what [the parties] knew at policy inception to be Known Pollution Conditions.” Suppl. Opp’n at 4. But the ELI Policy defines' Known Pollution Conditions to be “all conditions specifically described in the Scope of Work Endorsement,” and it provides that “Known Pollution Conditions constitute all of the conditions that are deemed known to the Insureds for the purposes of this Policy.”

Steadfast has not borne its burden to show the parties’ course of performance modified the terms of the ELI Policy.

*975 V. CONCLUSION

LMI’s motions for summary judgment are GRANTED IN PART as follows:

(1) The “Consent Agreement between Lennar Mare Island, the City of Vallejo and the State of California, California Environmental Protection Agency, Department of Toxic Substances Control Concerning Mare Island Naval Shipyard, Vallejo, CA,” dated April 16, 2001, constitutes “Governmental Authority”;

(2) For purposes of the ELI Policy, a person’s or entity’s actual knowledge of a pollution condition is insufficient to establish that condition is a Known Pollution Condition; and

(3) In all other respects, the motions are DENIED.

This order resolves ECF Nos. 160 and 186.

IT IS SO ORDERED.

.At the same hearing the parties offered arguments on Steadfast’s motion to amend its counterclaim. CCI appeared in opposition to that motion, but took no position on LMI’s current motions for summary judgment.

. See Lennar Mare Island, LLC v. Steadfast Ins. Co., No. 12-02182, 2014 WL 813140 (E.D.Cal. Feb. 28, 2014), recons. denied, 2014 WL 2002204 (E.D.Cal. May 15, 2014).

. See Lennar Mare Island, LLC v. Steadfast Ins. Co., 105 F.Supp.3d 1100 (E.D.Cal.2015).

. The ELI policy reads in relevant part as follows:

LMI Evid. in Support of Mot. Summ. J. Re Gov’t Auth. (Auth. Evid.), Ex. 1, at SJCE 167596 (bold typeface omitted); see also id. at SJCE 167554, 167599, 167600 (defining "First Named Insured” and "The Company").

. See Lennar Mare Island, LLC v. Steadfast Ins. Co., No. 12-02182, 2016 WL 829210 (E.D.Cal. Mar. 3, 2016).

. In their arguments the parties do not attribute meaning to bold typeface, which appears to indicate that a term is defined elsewhere in the policy. See Poll. Mem. at 6 n.4. The court's decision here does not turn on typeface, bold or otherwise.

. See Auth. Evid. Ex. 5, ¶ 1.1.

. “Site” is a defined term and generally refers to Mare Island. See Auth. Evid. Ex. 5, ¶ 1.2.

. This argument was absent from Steadfast's briefing. In most circumstances, a party abandons an argument by omitting it from an opposition brief. See, e.g., Stickling Pensioen fonds ABP v. Countrywide Fin. Corp., 802 F.Supp.2d 1125, 1132 (C.D.Cal.2011). Nevertheless, in its discretion, the court addresses the argument briefly.

. All citations to and quotations of that definition below are from this same page range. Bold typeface is omitted, as it does not communicate a meaning other than that the bold-ed term is defined. See supra note 6.

. The list in the policy is separated by both periods and semicolons, as reproduced above. This punctuation scheme, although perhaps an error, is not material to the court's decision here.

. Another form is a "course of dealing.” Cal. Civ. Proc. Code § 1856 (c). Course-of-dealing evidence and course-of-performance evidence are distinct. " 'Course of dealing/ ... is restricted, literally, to a sequence of conduct between the parties previous to the agreement. A sequence of conduct after or under the agreement, however, is a 'course of performance.’ " Cal. Com. Code § 1303 U.C.C. cmt. 2.

. In a previous order, the court signaled its suspicion this question may be resolved in Steadfast’s favor. See Order May 15, 2014, at 8-9. But as explained in the text above, after further review of the ELI policy and the definition of Known Pollution Conditions, the court now reaches the opposite conclusion.

. This exhibit appears not to have been hied on the CM/ECF system, but was included in the courtesy copies sent to the court by mail. A copy is filed concurrently with this order.

. Because the court concludes substantive California law precludes admission of Steadfast’s evidence, it does not reach LMI’s objections under the Federal Rules of Evidence.