ORDER RE: SUMMARY JUDGMENT

Re: ECF Nos. 54, 57, 59

Before the Court are Plaintiffs’ Motion for Summary Judgment, ECF No. 54, Federal Defendants’ Cross Motion for Summary Judgment, ECF No. 57, and Defendant Intervenors’ Cross Motion for Summary Judgment, ECF No. 59. For the reasons set forth below, the Court denies Defendants’ and Intervenors’ motions, and denies in part and grants in part Plaintiffs’ motion.

1. FACTUAL BACKGROUND

This case concerns the coastal marten. The marten is a small mammal in the weasel family with distinctive orange and yellow chest fur and a large bushy tail. MAR 1 20983. Coastal martens 2 inhabit old-growth forests west of the Rocky *1275 Mountains in Northern California, Southern Oregon, and Central Oregon. MAR 20985. Historically, the coastal marten lived “in all coastal Oregon counties and the coastal northern counties of California within late-successional coniferous forests.” MAR 22029. Fur trapping and logging in the mid to late 1900s, among other things, seriously impacted the coastal marten. MAR 13210, 3Í871. Today, the coastal marten population is comparatively small and reduced in distribution; only three populations of coastal marten remain: one in northern California, one in southern Oregon, and one in central Oregon. MAR 20999; MAR 20919 (explaining that the coastal marten population in California occupies “less than five percent of the historical range” of the marten in California); MAR 20989 (finding that the two coastal marten populations in Oregon “occupy approximately 15% of the historical range” of the marten in Oregon). The precise extent of that decline is the subject of this lawsuit.

On September 28,: 2010, the Center for Biological Diversity (“CBD”) and the Environmental Protection Information Center (“EPIC”) submitted a petition to the United States Fish and Wildlife Service (the “Service”), requesting that the coastal marten. be listed as an endangered or threatened species. MAR 22022. 3 On January 12, 2012, the Service published its initial 90-day finding. Id (referred to as the “90-day finding” or “initial finding”). The Service concluded that Plaintiffs’ petition “presented substantial information indicating that listing [of the coastal marten] may be warranted.” Id. The initial finding briefly described the three known coastal marten populations. First, the report identified a coastal Northern California marten population that was “estimated ■ to have undergone a 42 percent decline in occupancy” between 2001 and 2008 and “likely consists of fewer than 100 individuals.” MAR 1945. Second, the initial finding described “two disjunct” Oregon popula *1276 tions — “one in central coastal Oregon and one in southern coastal Oregon — both of which are believed to be in decline based mainly on a reduction in the number of martens trapped and anecdotal observations over time.” Id.

As a result of the 90-day finding, the Service initiated a comprehensive status review of the coastal marten. A draft report was first circulated on November 4, 2014, MAR 9242, and discussed in a team meeting on November 14, 2014, MAR 9567. According to minutes taken during the meeting, the majority of those in attendance preliminarily recommended listing the marten as endangered. MAR 10150-52 (nine recommended listing, one recommended listing only the California population, and two recommended against listing). One biologist, however, later sent a follow-up email objecting that the minutes made her “ ‘preliminary recommendation’ sound far more definitive than it was.” MAR 10165. 4 Discussion among the team continued throughout 2014 and early 2015, with at least two Oregon biologists expressing doubts about whether sufficient data existed to support listing. MAR 13354. By January 20, 2015, several members of the marten team agreed that another meeting was needed to “iron out the differences” within the core team. MAR 13330. Nevertheless, on January 23, 2015, the Pacific Southwest Regional Director issued a memorandum that recommended against listing the coastal marten. MAR 13522.

Consistent with this recommendation, the Service posted its 12-month finding on March 31, 2015, MAR 21030, and simultaneously released its underlying Species Report, MAR 20887. The Service stated that both the 12-month finding and the Species Report were based on the “best available scientific and commercial information” about the coastal marten. MAR 22023. 5 The 12-month finding, which found that the coastal marten was “not warranted for listing at this time,” was published in the Federal Register on April 7, 2015. MAR 22022. The 12-month finding first confirmed that the northern California, southern Oregon, and central Oregon marten populations comprise a single, listable entity, also known as a discrete population segment (“DPS”). MAR 22024. As required by DPS Policy, the Service determined that the coastal marten is both discrete (“separated from other populations of the same taxon as a consequence of physical, physiological, ecological, or behavioral factors”) and significant (of “biological and ecological importance ... to the taxon to which it belongs”). MAR 22026. The 12-month finding began by acknowledging that the coastal marten “were once considered relatively abundant,” but that “historical fur trapping is thought to have resulted in a significant contraction of coastal marten distribution and the extirpation of coastal marten from large portions of its historical range.” MAR 22029. Moreover, the Service stated that “much of the coastal marten’s historical habitat has been lost.” Id.

With this background in mind, the Service analyzed the five factors enumerated *1277 in section 4 of the Endangered Species Act (“ESA” or “the Act”), 16 U.S.C. 1533, for deciding whether to add a species to the Federal Lists of Endangered and Threatened Wildlife and Plants:

At bottom, the Service concluded that none of the stressors individually or together rose to the level of a threat to the coastal marten. MAR 22045.

The Service did not stop there, however, because under the ESA and the Service’s implementing regulations, a species may warrant listing if it is an endangered or a threatened species throughout all or a significant portion of its range.” MAR 22049 (emphasis added) (referred to as the “SPR analysis”). In other words, once it found that the coastal marten population as a whole did not warrant listing, the Service was still required to consider whether one of the three regional marten sub-populations (northern California, southern Oregon, or central Oregon) was uniquely vulnerable to extinction. The Service answered this question in the negative, after it concluded that “the overall level of stressors is not geographically concentrated in one portion of the coastal marten’s range.” MAR 22051.

II. PROCEDURAL BACKGROUND

Plaintiffs filed their Complaint challenging the “not warranted” finding on December 15, 2015. ECF No. 1. On February 18, 2016, Siskiyou County, California, Douglas County, Oregon, the American Forest Resource Council, the National Association of Home Builders, the California Forestry Association, the Oregon Forest Industries Council, and Douglas Timber Operators (“the Intervenors”), moved to intervene. ECF No. 13. Plaintiffs did not oppose and the Court granted the Intervenors’ motion on March 7, 2016. ECF No. 49.

On August 17, 2016, Plaintiffs moved for summary judgment. ECF No. 54. Plaintiffs argue that two main components of the Service’s 12-month finding are not supported, as required, by the “best scientific and commercial data available.” First, Plaintiffs claim that the Service erred in its analysis of the fifth ESA factor when it concluded that “small and isolated population effects” did not rise to the level of a threat to the coastal marten. MAR 22043. Plaintiffs argue that the best available evidence demonstrates that the coastal marten populations are “small and declining” and that they are “functionally isolated.” ECF No. 54 at 17-20. Second, Plaintiffs challenge the Service’s conclusion that the coastal marten is not threatened or endangered throughout any significant population of its range. Id. at 21; MAR 22051. They claim that the best available evidence shows that the overall level of stressors is “concentrated in the California potion of the coastal marten’s range.” Id.

In October 2016, Defendants and Inter-venors each filed cross motions for summary judgment. ECF Nos. 57, 59. Both motions argue that the Service properly analyzed the five ESA factors, that the not warranted finding is supported by the best available evidence, and that this Court *1278 must therefore defer to the Service’s decision-. 6

III. LEGAL STANDARD 7

This Court’s review of an agency’s. compliance with the ESA is governed by the Administrative Procedure Act (“APA”). Native Ecosystems Council v. Dombeck, 304 F.3d 886, 901 (9th Cir. 2002). Under the APA, a court may invalidate only those agency actions found to be “arbitrary, capricious, an abuse of discretion, or otherwise5 not in accordance with, law.” - 5 U.S.C. § 706 (2)(A). This is a “deferential standard,” and the Court’s job “is simply to ensure that the agency considered the relevant factors and articulated a rational connection between the facts found and the choices made,” Nw. Ecosystem Alliance, 475 F.3d at 1140 (internal quotation marks omitted); see also Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 , 103 S.Ct. 2856 , 77 L.Ed.2d 443 (1983). (“[A]n agency rule would be arbitrary and capricious if.the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision, that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”).

IV. ANALYSIS

A. The Service’s Not Warranted Finding

Plaintiffs do not appear to challenge the Service’s analysis of the first four ESA listing factors, instead limiting their arguments to factor five. The. Court, agrees with Defendants that the Service’s, conclusions regarding the first four factors of the 12-month finding are supported by the best available evidence and are entitled to deference. : .

For the first factor, “the present or threatened destruction, modification, or curtailment of- its habitat or range,” the Service examined the risk from wildfire, climate change, vegetation management, and development. MAR 22035-38, Particularly because much of the marten’s suitable habitat within the three population areas is on federal lands and protected under the , Northwest Forest Plan (“NWFP”), the Service concluded that factor one did not rise tp the level of a threat. As for the second factor, “[o]vqrutilization for commercial, recreational, scientific, or educational purposes,” the Service noted that, historically, unregulated fur trapping presented the main threat to the coastal marten, but concluded that trapping had been eliminated in California and reduced in Oregon. MAR 22038. Third, the Service analyzed the risk to the marten from dis *1279 ease or predation and found no evidence that either rose to the level of a threat. MAR 22048. Looking to the fourth factor, the “inadequacy of existing regulatory mechanisms,” the Service determined that, far from posing a threat, multiple federal land use plans and state regulations may help “abate the large-scale loss of forested habitat-types deemed essential for coastal martens.” Id. The Court agrees with Defendants that, at least as to these four factors, the Service “considered the relevant factors and articulated a rational connection between the facts found and the choices made.” Nw. Ecosystem Alliance, 475 F.3d at 1140.

The parties dispute, however, whether the best available evidence supports the Service’s analysis of factor five: “Other natural or manmade factors affecting [the marten’s] continued existence.” Specifically, the Service concluded that the coastal marten population is not small enough or isolated enough to pose a threat to its existence. Plaintiffs dispute both findings. 8

1. Population Size

Plaintiffs argue that coastal marten populations are “small and declining,” not “robust.” EOF No. 54 at 6. 9 The Service’s contrary finding, Plaintiffs claim, is inconsistent with the best available evidence and therefore contrary to law. For the California population, Plaintiffs emphasize that the Species Report documented a 42% population decline between 2001 and 2008 and a total population of less than 100 individuals, 10 MAR 20928-29. The Service responds that, although the California population may be small, no evidence indicates that it continued to decline after 2008. It points to the Species Report’s statement that “[p]reliminary occupancy estimates for [] 2012 sampling were similar to results from 2008 ... suggesting that marten population abundance in northern coastal California did not change over that time period.” MAR 20928-29. Plaintiffs dispute the Service’s characterization of the population as stabilized. They point to a comment by a team biologist on a draft of the 12-month finding that echoes their argument:

MAR 18424.

Plaintiffs similarly claim that the best ayailable evidence shows the Oregon population to be small and declining. They again point to the Species Report, which acknowledges that “[tjhere are no known population size estimates for the two ex *1280 tant populations in coastal Oregon, but researchers consider both populations likely to be relatively small.” MAR 20933. A table of marten abundance trends contained in the Report states that “[b]oth [Oregon] populations are believed to be in decline based mainly on a reduction in the number of martens legally trapped and anecdotal sightings over time.” MAR 20930. Plaintiffs also point to the Species Report’s finding that an absence of reported marten road kills “suggest[s] low numbers of martens.” MAR 20928.

Defendants focus principally on the lack of survey data on Oregon population size. See, e.g„ MAR 20943 (“[CJoastal marten distributional data and home range size estimates within [southern and central Oregon] are not available”). They note findings that coastal “martens are most common in coastal Oregon,” MAR 36510, that “suitable conditions exist” to support a marten population in Oregon, MAR 21896, and that current and ongoing survey efforts quickly detected marten presence, MAR 22043. Defendants also challenge the road kill data as relating only to Northern Oregon and Southern Washington, and indeed the underlying study contains the following finding:

MAR 14019. This directly contradicts the Species Report’s characterization of the study as showing “an absence of reported marten road kills long coastal U.S. Highway 101 in northern coastal Oregon, southern coastal Oregon, and northern coastal California” MAR 20928 (emphasis added).

“The ESA instructs the Service to make its determinations ‘solely on the basis of the best scientific and commercial data available,’ and the Service may not ignore evidence simply because it falls short of absolute scientific certainty.” Nw. Ecosystem All., 475 F.3d at 1145. Here, Defendants are correct that the Service lacked the most recent data on population size in California and clear estimates in the diffieult-to-survey Oregon habitats. But “[i]f the science on population size and trends is underdeveloped and unclear, the Secretary cannot reasonably infer that the absence of evidence of population decline equates to evidence of persistence.” Tucson Herpetological Soc. v. Salazar, 566 F.3d 870, 879 (9th Cir. 2009).

The Court agrees with Plaintiffs that the best evidence available requires the conclusion that the California marten population is small and declining. The Service does not dispute that the California marten population declined precipitously in the 2000s, falling to a population of less than 100 animals. MAR 20928-29. In other words, there is no doubt that the California population is exceedingly small, and although 2012 survey results suggest similar numbers to 2008, one of the Service’s own biologists rejected the notion that those results demonstrated stabilization of the California marten population. MAR 18424.

Tucson is not “inapposite” or “misplaced” as Defendants and Intervenors claim. ECF No. 62 at 8, ECF No. 61 at 11. *1281 In that case, the Ninth Circuit deemed it unreasonable to “infer that the absence of evidence of [lizard] population decline equates to evidence of persistence” where the “science on population size and trends [wa]s underdeveloped and unclear.” Tucson, 566 F.3d at 879 . That is precisely what the Service did here. In the face of evidence of small and declining marten populations in California that was outdated by a few years, the Service relied on the absence of the most recent data to conclude that the California population was stable and that its size therefore did not pose a threat. Tucson prohibits that approach. Defendants claim they never argued that “coastal marten populations are persistent or increasing, but simply [ ] that there is no data of current population declines.” ECF No. 61 at 12. But these statements are just two sides of the same coin; what else would the Service be trying to show from the absence of data on current population declines if not persistence of the marten population? It is true that Tucscon also held that “[t]he absence of conclusive evidence of persistence, standing alone, without persuasive evidence of widespread decline, may not be enough to establish that the Secretary must list the lizard as threatened or endangered.” Id. (emphasis added). But Plaintiffs are not asking the Court to direct the Service to list the coastal marten; rather, they ask for an order setting aside the 12-month finding and requiring the Service to prepare a new one. ECF No. 54 at 22.

Perhaps recognizing the lack of evidence demonstrating stable marten population size, the Service focuses the Court’s attention on the regulatory mechanisms in place that will help preserve the coastal marten’s habitat. The Court does not disagree that the Service properly weighed regulatory mechanisms (the fourth factor) against listing, but that does not change the Court’s conclusion that the Service failed to 'recognize that the best available evidence demonstrated small and declining marten populations. If Plaintiffs were challenging how the Service weighed the marten’s small and declining population size against the other ESA factors, that would be a different case with perhaps a different outcome. But here the problem is that the Service failed to recognize that the evidence showed a small and declining California marten population in the first place. That failure requires the Court to set aside, the 12-month finding.

The same is not true of Oregon. At the time the 12-Month Finding was prepared, high quality surveys of marten abundance were unavailable and no data clearly showed population decline. Unlike in California, where studies between 2001 and 2008 documented a 42 percent decline in occupancy, MAR 1945, no similar survey data had been collected related to the Oregon marten populations, MAR 22044. Also distinguishing Oregon from' California is the fact that recent (and still ongoing) surveys in Oregon have quickly' detected the presence of coastal marten. MAR 22044; MAR 29766 (detecting two martens within the first week). One report by several marten experts summarized the situation in Oregon as follow: Although not all evidence is as convincing as the results of detection surveys, the data suggest that of the 3 states, martens are most common-in coastal Oregon.” MAR 36510.

In response, Plaintiffs argue that trapping data,' anecdotal sighting reports, and road kill data demonstrate that the Oregon population is indeed small and declining. Although trapping data and anecdotal sighting reports may indicate low numbers of martens in Oregon, MAR 20928-30, the *1282 road kill data analyzed by marten experts actually suggested the opposite. The Species Report states that:

MAR 20928. This misstates the findings of the underlying Zielinski study. In fact, the study found that an “absence of reported road kills along coastal Highway 101 in northern Oregon and southern Washington, in contrast to the dozen or so on the same highway in central Oregon, also suggests low numbers of martens,” MAR 36509-10. In other words, the road kill data indicates that the Oregon marten population is relatively large, not small. Nor is the trapping data particularly helpful to Plaintiffs, since it showed only that “number of -martens harvested in coastal Oregon counties has declined since the 1940s.” MAR 36505. That says very little about the Oregon marten’s current, population size.

Certainly, “a rigorous, large-scale study of [Oregon marten population size] would be preferable, [but] in the absence of such a study, credible anecdotal evidence represents the ‘best scientific ... data available’ and cannot be ignored.” Nw. Ecosystem All., 475 F.3d at 1147 (emphasis in original) (quoting 16 U.S.C. § 1533 (b)((1)(A))). This Court may not second guess the Service’s conclusion, particularly given that most of the evidence suggested relatively stable marten abundance levels in central and southern Oregon. .

In sum, the Court finds that the best available evidence supported the Service’s conclusions about the size of the Oregon marten populations, but not the California population.

2. Population Isolation

Plaintiffs also argue that Service erred when it concluded that the three coastal marten populations were not functionally isolated. ECF No. 54 at 19. According to the Species Report, the three marten populations “are separated from one another by distances 2 to 4 times greater than juvenile dispersal distance reported for American martens ... suggesting that all three extant populations are' functionally isolated from one another.” MAR 20933. The Report also explains that-“each of the three extant coastal marten populations are likely to be functionally isolated from one another based either on marten survey results or limited habitat availability between , tlie populations.” MAR 20989-90. Commenting on an earlier draft of the 12-month finding, one team member put it more starkly: “Any suggestion in 'this document that the three popu-' lations are not isolated is inaccurate and not supported by the literature or any other available source of information.” MAR 18403. The Species Report described recent studies that “confirmed” that the southern. Oregon and northern California populations are “functionally isolated from one another.” MAR 20933. These findings were apparently based on three years of data on the California marten population at over 3,000 survey locations where tracked animals did not move between the *1283 two areas. MAR 18120, MAR 20919. Moreover, the likelihood of detecting dispersal in this survey was estimated at 90%. MAR 22044. Regarding the two Oregon populations, the Species Report noted that “[l]ess survey effort between the two coastal Oregon populations makes it more difficult, to determine whether those populations are isolated from each other, although the limited amount of suitable habitat between the two populations suggests they are.” MAR 20933-34.

Defendants offer several responses. First, they argue that the maximum known dispersal of á juvenile marten is 43 miles, which falls within the range of the three populations. MAR 18132. Further, studies of martens in Canada recorded certain instances of adult martens traveling well beyond their mean dispersal ranges when faced with starvation or when lost and in search of a new home. MAR 32574, MAR 31977. 11 Defendants point to one table in the Species Reports that shows what appear to be two marten detections between the northern California and southern Oregon populations over an uncertain time period. MAR 20924. There is no evidence that these detections reflected movement between the two populations, however. Indeed, the only studiés tracking collared martens in California did not document any movement from California to the southern Oregon range. MAR 18120. Defendants also argue that the Service’s conclusion is supported by the existence of suitable habitat between the northern California and southern Oregon populations, MAR 22043, and “that management of federal lands under the NWFP should contribute to improved connectivity,” MAR 22044. As an initial matter, this argument has little bearing on the connectivity of the central and southern Oregon populations, which the Species Report found to enjoy “little habitat connectivity.” MAR 201916. 12 Nor is the NWFP analogous to the conservation plans that the Ninth Circuit found “highly significant” in Greater Yellowstone Coal., Inc. v. Servheen, 665 F.3d 1015, 1031 (9th Cir. 2011). There, the land management plans included provisions that directly related to the animal at issue in the case, the grizzly bear, making it reasonable to rely on those provisions to protect the bear if it were de-listed. Moreover, the grizzly bear population had “rebounded” due to those conservation efforts! Id. The same is not true here, where the Service does not suggest that the NWFP already contains special protections for the coastal marten, nor that those protections (which have been in place for several decades) have led to a resurgence of its population numbers. 13 Finally, the Service argues that a Canadian study showing minimal “genetic differenti *1284 ation” between “martens in fragmented landscapes ... is indicative of great dispersal capacity in martens, and their results suggest that a few successful long-distance dispersers create enough gene” to mitigate the effects of isolation. MAR 220444; MAR 22659.

Again, the Court agrees with Plaintiffs that the Service’s finding that the coastal marten populations are not functionally isolated is not supported by the best available evidence. To start, the 12-month finding directly contradicts the Service Report’s conclusion that “each of the three extant coastal marten populations [is] likely to be functionally isolated from one another based either on marten survey results or limited habitat availability between the populations.” MAR 20989-90. This stark inconsistency makes the 12-month finding appear arbitrary. While the Service “is not required to accept the [Species Report’s] conclusion,” it “must use the [Report’s] data in reaching its listing decision.” Kern Cty. Farm Bureau v. Allen, 450 F.3d 1072 , 1080 (9th Cir. 2006).

Moreover, although the Service is entitled to deference when weighing conflicting views or studies, it is unreasonable to disregard a three-year study in one of the relevant population areas with over 3,000 locations where not a single tracked animal moved from northern California to southern Oregon. MAR 18120, MAR 20919. Indeed, this study bolsters the Species Report’s statement that the coastal marten populations “are separated from one another by distances 2 to 4 times greater than juvenile dispersal distance reported for American martens ... suggesting that all three extant populations are functionally isolated from one another.” MAR 20933. The Service offers no convincing rationale for relying instead on 1) isolated examples of Canadian martens dispersing beyond their mean distances, 14 2) the fact that low quality habitat between California and southern Oregon could theoretically allow for dispersal, and 3) genetic differentiation studies of martens in Canada. “The purpose of the best available science standard is to prevent an agency from basing its action on speculation and surmise. Under this standard, an agency must not disregard available scientific evidence that is in some way better than the evidence it relies on.” San Luis & Delta-Mendota Water Auth. v. Locke, 776 F.3d 971, 995 (9th Cir. 2014). Evidence of the mean dispersal range of the coastal marten population and the fact that no instances of dispersal between the three populations have been documented is more direct evidence on the question of isolation than two studies on a different species of marten in Canada and the existence of poor quality connective habitat that could, in the abstract, allow for dispersal. 15 In other *1285 words, the Service improperly based its isolation finding on “speculation,” rather than on the data to which it had access. In so finding, the Court is fillfilling its role to ensure that the Service does not “disregard available scientific evidence that is in some way better than the evidence it relies on;” it is not “choos[ing] among scientific studies ... and ordering] the agency to explain every possible scientific uncertainty.” The Lands Council v. McNair, 537 F.3d 981, 988 (9th Cir. 2008) (en banc).

Plaintiffs’ argument is bolstered by the Service’s own DPS analysis. In analyzing the “discreteness” of the coastal marten, the 12-month finding explained:

MAR 22025. The Service used this data to conclude that the coastal marten was isolated from other Pacific martens further east. While the Service certainly relied on other factors to make its final decision that the coastal marten was a DPS, it is still notable that the Service seems to discount this same evidence when analyzing whether the three coastal marten populations are isolated from one another.

In sum, the Court finds that the best available evidence does not support the Service’s determination that the three coastal marten populations are not functionally isolated. As above, this error does not mean that the Service must list the marten as endangered. It means the Service could not have properly weighed factor five against the other four factors outlined in the ESA because it failed to acknowledge the small, declining, and isolated nature of the coastal marten populations. Therefore, its 12-month finding is flawed and must be reconsidered.

B. SPR Analysis

Plaintiffs’ final challenge to the 12-month finding relates to its SPR analysis. Specifically, Plaintiffs dispute the Service’s conclusion that “the overall level of stres-sors is not geographically concentrated in one portion of the coastal marten’s range.” MAR 22051. They claim that the best available evidence shows that stressors are concentrated in the California portion of the coastal marten’s range. ECF No. 54 at *1286 21. In support of this argument, Plaintiffs note that the Species Report found that habitat loss from wildfire and climate change “will likely impact the CA_EPA and SCOJEPA [i.e., the California and southern coastal Oregon] populations .to a much greater extent than the CCO-EPA [i.e., central coastal Oregon] population.” MAR 20994-95. .The Species Report also documents higher rodenticide exposure in California than for the two Oregon populations. MAR 20999. Overall, therefore, the Report concluded that “[t]he population in northern coastal California is especially vulnerable to extirpation due to its small size and very limited distribution, its location within a fire-prone area, and the large number of illegal marijuana grows that likely occur within occupied suitable habitat potentially exposing1 martens and marten prey to anticoagulant rodenticidés.'” MAR 21000.

■ Defendants offer persuasive responses to this argument. First, they point out that the wildfire and climate change stressors are not unique to California; the Species Report specifically states that both will impact the California mid southern coastal Oregon populations. MAR 20994-95. The Service also explained that the relatively wet and foggy marten habitat reduced the risk of fire on the marten’s habitat. MAR 22045. As for rodenticide exposure, the Service ■ acknowledged that grow sites “may possibly occur to a greater extent” in California than Oregon, but explained that no data supported a finding that toxicants are having population effects on the coastal marten. MAR 22051. In fact, only one positive exposure record for a coastal marten has been documented. Id. And although Plaintiffs criticize the service for hot explicitly using the term “balancing” in its assessment of these and other stressors among the three coastal marten populations, the 12-month finding clearly analyzes each factor in the context of the others. See MAR 22049-51.

The problem, however, is that the Service’s erroneous conclusion about the California population’s size, discussed above, surely influenced its SPR analysis. Even if the Service sufficiently explained away the wildfire, climate change, and ro-dehticide risks, if it failed to recognize the California population’s isolation and. small and declining numbers, it could not have properly assessed whether that sub-population was, as the Species Report found, “especially vulnerable to extirpation.” The Court directs the Service to reconsider its SPR analysis in light of its revised analysis of the five ESA factors.

CONCLUSION

■The. .Court denies Defendants’ and Inter-venors’ motions for summary judgment. With .the exception of their challenge to the Service’s finding that the Oregon marten population is not small and declining, the Court grants Plaintiffs’ motion for summary judgment.

IT IS SO ORDERED.

. MAR refers to the Administrative Record, which was lodged with the Court at ECF- No. 52.

. Plaintiffs’ petition also resulted in a change to the taxonomy of the marten. Before, coastal martens were divided into two subspecies: *1275 the humboldtensis in coastal northern California, and the caurina in coastal Oregon. After reviewing recent studies that the two subspecies actually represented a “single evolutionary entity,” however, the Service in its 12-month finding concluded that the coastal martens in Oregon and California comprise a "distinct population segment,” or "DPS.” MAR 22026. As explained below, the Service then declined to list this DPS as endangered. In this order, the Court refers to the DPS at issue as the "coastal marten.” • '

. The Ninth Circuit offered the following summary of the Service's listing process: ■

Nw. Ecosystem All. v. U.S. Fish & Wildlife Serv., 475 F.3d 1136 , 1137-38 (9th Cir. 2007).

. This team member, Dr. Michele Merola Zwartjes, appears to have been the main opponent of listing. Throughout their briefs, Defendants argue that the team vigorously debated the stressors facing the marten. But, in most cases, the only team member they point to is Dr. Zwartjes.

. This "best available evidence” requirement is set forth in the ESA itself. 16 U.S.C. § 1553 (b)(1)(A).

. Because the briefing by Defendants and In-tervenors is similar, the Court refers to their arguments jointly throughout this order.

, Before seeking summary judgment, an ESA plaintiff must establish standing by setting forth "by affidavit or other evidence ‘specific facts,’ which for- purposes of Summary judgment motion will be taken to be true.” Biodiversity Legal Found. v. Badgley, 309 F.3d 1166 , 1178 (9th Cir. 2002) (quoting Lujan v. Defenders of Wildlife, 504 U.S, 555, 561, 112 S.Ct. 2130 , 119 L.Ed.2d 351 (1992)). Here, Plaintiffs have submitted two sworn declarations by members of Plaintiffs' organizations describing the harm they have suffered from the Service's decision not to list the coastal marten. ECF Nos. 54-1, 54-2. Defendants do not contest Plaintiffs' standing and the Court find's that it has been established for purposes of summary judgment.

. The parties agree that, as a general matter, "small, isolated populations are vulnerable to extinction due to genetic problems, demographic and environmental fluctuations, and natural catastrophes.” MAR 20989. The only question is whether the coastal marten population is sufficiently small and isolated to suffer those adverse impacts.

. The Service has not claimed that the coastal marten population is "robust.”

.Intervenors' response that "the 2008 and 2012 surveys were designed in such a way that they have limited ability to detect increases in occupancy” is not particularly persuasive given that the 100-individual estimate doubled the number of martens detected in those studies to account for that limitation.

. Adult martens, however, are less likely than juveniles to leave their established home ranges. MAR 20900.

. Dr. Zwartjes argued that "little” is not the same as "none,” MAR 18132, but it is clear that the suitable habitat that does exist is "low quality," which tends to significantly reduce the dispersal rates and distances of martens, MAR 20913-15.

. The Court does not underestimate the positive impacts of the NWFP, which placed millions of acres of marten habitat into reserves, and, agrees that it was properly weighed against listing under Factor D. The Service overstated its importance here, however, particularly since NWFP protections have slowed the “rate of loss of marten habitat federal lands due to timber harvest” but likely not the "overall decline in late-supcessional and old-growth forest.” MAR 14387. Development is still allowed on federal “matrix lands.” MAR 22040. Moreover, only 31% of the marten’s habitat consists of federal lands in the first place. Id.

. Defendants imply that information about these Canadian outliers was purposefully excluded from the Species Report, pointing to an email from Dr. Zwartjes that commented to the disagreeing biologist, "[p]erhaps, though, there is a reason that you did not include that information?” MAR 18118. The Court declines to draw this inference. The data could just have easily been excluded because the biologist thought it less relevant than the California survey, which examined the coastal marten population that was the subject of the Report.

. Defendants emphasize that two biologists working on the Species Report disagreed precisely on this point. See MAR 18118-18127. Arguing against a finding of functional isolation, Dr. Zwartjes focused on the Canadian studies and limited habitat connectivity between the populations, while Dr. Gregory Schmidt relied on the mean maximum dispersal for juvenile martens and the recent study *1285 tracking collared animals to argue the opposite. The existence of this debate, Defendants say, suggests a contested scientific question whose ultimate resolution merits increased deference to the agency. ECF No. 57 at 19. But as Dr. Zwartjes recognized, Dr. Schmidt was the “the species expert here,” and yet the 12-month finding adopts Dr. Zwartjes’s conclusions without addressing the concerns raised by Dr. Schmidt. “An agency complies with the best available science standard so long as it does not ignore available studies, even if it disagrees with or discredits them.” San Luis & Delta-Mendota Water Auth. v. Locke, 776 F.3d 971, 995 (9th Cir. 2014). This debate between the two biologists serves to illuminate that failure here.