Plaintiffs appeal the district court’s denial of their request for a preliminary injunction to prevent the drilling of certain oil and gas wells in the Mancos Shale formation of the San Juan Basin in New Mexico. The district court concluded that Plaintiffs had failed to satisfy three of the four elements required to obtain a preliminary injunction: (1) Plaintiffs had not demonstrated a substantial likelihood of success on the merits of their claims; (2) the balance of harms weighed against Plaintiffs; and (3) Plaintiffs failed to show that the public interest favored an injunction. Plaintiffs challenge each of these conclusions on appeal.
I.
The San Juan Basin is a large geographic region in the southwestern United States, including part of New Mexico. Drilling for oil and gas has occurred in the Basin for more than sixty years, .and the Basin is currently one of the most prolific sources of natural gas in the country. The Basin includes both public and private lands. Many of the public lands and resources fall under the jurisdiction of the Bureau of Land Management’s Farming-ton Field Office in New Mexico, which manages these lands and resources under its published Resource Management Plan.
In 2000, the BLM initiated the process of revising its existing RMP, which had been published in 1988. As part of this process, the BLM contracted with the New Mexico Institute of Mining and Geology to develop a “reasonably foreseeable development scenario,” or RFDS, to predict the foreseeable oil and gas development likely to occur over the next twenty years. Based on historic production data and available geologic and engineering evidence, the RFDS estimated that 9,970 new oil and gas wells would be drilled on federally managed lands in the New Mexico portion of the San Juan Basin during this time period. Of these wells, the RFDS estimated that more than forty percent would be “Dakota, Mancos” gas wells— wells that could produce gas from both the Mancos geologic horizon and the Dakota geologic horizon that lies below it. (Appel-lees’ Supplemental App. at 37.) The RFDS estimated that only 180 new oil wells would be drilled in the Mancos Shale, due to the fact that most reservoirs in the Mancos Shale were approaching depletion under then-current technologies, but it noted that “there is excellent potential for the Mancos to be further evaluated.” (Joint App. at 182.)
In 2003, the BLM issued its Proposed Resource Management Plan and Final Environmental Impact Statement (PRMP/FEIS). In this document, the BLM referred to the predictions and analysis contained in the RFDS in order to assess four proposed alternatives for managing federal lands in the San Juan Basin, including the “balanced approach” the agency ultimately decided to adopt. {Id. at 499.) Under this balanced approach, the BLM analyzed the cumulative impacts of an estimated 9,942 new wells in the San Juan Basin—approximately the same num*1280ber predicted in the 2001 RFDS—by looking at, for instance, the likely air quality impacts from the drilling and operation of this many new wells in the region. The PRMP/FEIS did not discuss specific sites or approve any individual wells, although it assumed the majority of new wells would be drilled in the high development area in the northern part of the managed area. The BLM issued its final RMP, adopting the Alternative D balanced approach, in December 2003.
Although this RMP generally allows for drilling in the San Juan Basin, a new well may not be drilled unless the operator first receives specific approval of its drilling plan through the submission of an application for a permit to drill, or APD, to the BLM. When the agency receives an APD, it reviews the planned drilling activity and prepares an environmental assessment of the environmental impacts of the proposed drilling, after which the agency may approve the APD as submitted, approve it with appropriate modifications or conditions, or deny it. In preparing this environmental assessment, the agency may incorporate by reference the general but more comprehensive environmental analysis included within its published RMP and focus the APD’s environmental assessment “on the issues specific to the [proposed drilling].” 40 C.F.R. § 1502.20. This process is generally referred to as tiering. See id.
Beginning in about 2014, the BLM began receiving more APDs than anticipated for oil wells in the Mancos Shale. At the time the RMP was issued, oil and gas in the San Juan Basin were generally produced through hydraulic fracturing of vertical oil wells. However, subsequent technological advances have made it economical for oil and gas operators in the San Juan Basin to drill horizontal wells and conduct multi-stage fracturing, which enables them to tap into oil and gas reserves that are not readily accessible through traditional vertical drilling techniques and produce four times as much oil and gas per well as they could have produced from a single vertical well. Between 2010 and 2015, the BLM approved approximately 265 APDs for drilling in the Mancos Shale. In its environmental assessment of each APD or set of related APDs, the agency reviewed the site-specific and cumulative impacts likely to arise from the proposed drilling. In its assessment of some of these impacts, and particularly in its discussion of the cumulative impacts, the BLM referred to its broader but more comprehensive environmental analysis in the 2003 RMP.
In 2014, the BLM prepared a new RFDS to better predict the Mancos Shale’s potential for oil and gas development. The agency is now working on an RMP amendment to account for the estimated additional 1,930 oil wells and 2,000 gas wells in the Mancos Shale.
In March 2015, Plaintiffs filed this lawsuit under the National Environmental Policy Act, challenging 260 APDs in the Mancos Shale. Plaintiffs then moved for a preliminary injunction to prevent drilling on approved wells while this litigation is ongoing.1 The district court denied this motion in a 101-page decision, holding that Plaintiffs were not likely to succeed on the merits of their claims and that Plaintiffs’ potential harm was outweighed by the economic harms which the operators and the public would suffer if drilling was enjoined. This appeal followed.
*1281II.
We review the district court denial of a preliminary injunction for an abuse of discretion. Wilderness Workshop v. BLM, 531 F.3d 1220, 1223 (10th Cir. 2008). “An abuse of discretion occurs only when the trial court bases its decision on an erroneous conclusion of law or where there is no rational basis in the evidence for the ruling.” Id. at 1224 (internal quotation marks omitted). “Our review of the district court’s exercise of discretion is narrow, and the merits may be considered on appeal only insofar as they bear on the issue of judicial discretion.” Gen. Motors Corp. v. Urban Gorilla, LLC, 500 F.3d 1222, 1226 (10th Cir. 2007) (internal quotation marks, ellipses, and citation omitted).
“In order to receive a preliminary injunction, the plaintiff must establish the following factors: (1) a substantial likelihood of prevailing on the merits; (2) irreparable harm unless the injunction is issued; (3) that the threatened injury outweighs the harm' that the preliminary injunction may cause the opposing party; and- (4) that the injunction, if issued, will not adversely affect the public interest.” Davis v. Mineta, 302 F.3d 1104, 1111 (10th Cir. 2002) (internal quotation marks and brackets omitted). “Because a preliminary injunction is' an extraordinary remedy, the movant’s right to relief must be clear-and unequivocal.” Wilderness Workshop, 531 F.3d at 1224 (internal quotation marks and brackets omitted).
The district court concluded that Plaintiffs had shown irreparable harm but had not satisfied the other three prerequisites for obtaining a preliminary injunction: Plaintiffs had not shown that they were likely to succeed on the merits; Plaintiffs had not demonstrated that the threatened environmental harms outweighed the certain economic harms the operators would face if they were prevented from drilling; and the public interest weighed against the injunction. Because each of these elements is a prerequisite for obtaining a preliminary injunction, we will not reverse the district court’s denial of injunctive relief unless we are persuaded that the court abused its discretion as to all three issues, and if we see no abuse of discretion as to any one prerequisite, we must affirm the court’s ruling.
Plaintiffs raise four main arguments as to why the district court erred in concluding that they had not demonstrated a substantial likelihood of success on the merits of their claim: (1) the district court applied too high a standard of “likelihood” in evaluating their claims; (2) the court should have concluded that the new horizontal drilling and multi-stage fracturing technologies will lead to environmental harm of a greater magnitude than was anticipated in 2003, so it is arbitrary and capricious for the BLM to continue tiering its analysis to the 2003 RMP; (3) the court should have concluded that these new technologies will lead to environmental harms that are qualitatively different from those anticipated in 2003, so it is arbitrary and capricious for the agency to continue tiering to the 2003 RMP; and (4) the court should have concluded that the BLM’s continued approval of new wells will impermissibly limit the BLM’s options as it amends the 2003 RMP, thus violating the regulations and satisfying the arbitrary and capricious standard of agency review. We consider each of these arguments in turn.
Plaintiffs first argue the district court committed legal error in describing and applying the applicable test for determining whether a movant has shown a substantial likelihood of success on the merits of his claim. In Davis v. Mineta, we stated that when the other three requirements “tip strongly in [the plaintiff’s] favor, the test is modified, and the plaintiff may meet *1282the requirement for showing success on the merits by showing that questions going to the merits are so serious, substantial, difficult, and doubtful as to make the issue ripe for litigation and deserving of more deliberate investigation.” 302 F.3d at 1111 (internal quotation marks omitted). Plaintiffs argue that the district court erred in failing to apply this modified test.
However, as the district court correctly noted, our modified test is inconsistent with the Supreme Court’s recent decision in Winter v. Natural Resources Defense Council, 555 U.S. 7, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008). In Winter, the Court overruled the Ninth Circuit’s application of a modified preliminary injunction test under which plaintiffs who demonstrated a strong likelihood of prevailing on the merits could receive a preliminary injunction based only on a possibility, rather than a likelihood, of irreparable harm. The Court held that the Ninth Circuit had impermis-sibly deviated from the Supreme Court’s “frequently reiterated standard” for preliminary relief, which requires a showing of a likelihood of harm, and permitted a more lenient standard of relief “inconsistent with [the Court’s] characterization of injunctive relief as an extraordinary remedy that may only be awarded upon a clear showing that the plaintiff is entitled to such relief.” Id. at 22, 129 S.Ct. 365.
Although the Ninth Circuit standard overruled in Winter dealt with a different prong of the preliminary injunction test than the prong at issue in this case, Winter’s rationale seems to apply with equal force here. “We do not ... approach opinions of the Supreme Court with a view to reaching the narrowest construction possible. Instead, we have said that this court considers itself bound by Supreme Court dicta almost as firmly as by the Court’s outright holdings, particularly when the dicta is recent and not enfeebled by later statements.” United States v. Nelson, 383 F.3d 1227, 1232 (10th Cir. 2004) (internal quotation marks omitted); see also United States v. Brooks, 751 F.3d 1204, 1209 (10th Cir. 2014) (noting that our general rule against overturning a prior panel decision “does not apply. ... when the Supreme Court issues an intervening decision that is contrary to or invalidates our previous analysis” (internal quotation marks omitted)). Under Winter’s rationale, any modified test which relaxes one of the prongs for preliminary relief and thus deviates from the standard test is impermissible. We accordingly hold that the district court did not abuse, its discretion in simply applying the Supreme Court’s “frequently reiterated standard” for preliminary relief, including the requirement that the plaintiff must show he is likely to succeed on the merits.
Plaintiffs also argue the district court went even beyond the normal standard for preliminary relief by requiring them to meet an “elevated burden” that was closer to a certainty of success than a reasonable likelihood. (Appellants’ Opening Br. at 18.) This argument is based on the district court’s preliminary discussion of how a “reasonable likelihood of success” standard may be understood in the context of a bench trial where the district court is reviewing an agency decision under the arbitrary and capricious standard of review. However, Plaintiffs ignore the fact that the district court then explicitly stated it would “apply the standard as written” (Joint App. at 133), and we see no indication the district court actually deviated from this standard in evaluating the merits of Plaintiffs’ case.
Next, Plaintiffs argue they sufficiently demonstrated a likelihood of success on the merits because they presented evidence that the new horizontal drilling and multi-stage fracturing techniques be*1283ing applied in the San Juan Basin will lead to harm of a greater magnitude than was anticipated when the BLM drafted the 2003 RMP. These new drilling techniques have greatly increased access to oil and gas reserves that were not previously targeted for development and have given rise to much higher levels of development in the Mancos Shale than the BLM previously estimated and accounted for. Moreover, horizontal drilling and multi-stage fracturing may have greater environmental impacts than vertical drilling and older fracturing techniques. For instance, according to Plaintiffs, each horizontal well has approximately double the surface impact of each vertical well, and drilling a horizontal well leads to a 242%-333% increase in air pollutant emissions over the drilling of a vertical well. Plaintiffs argue that this increased level of development and environmental impacts means that the BLM acted arbitrarily and capriciously when it tiered its analysis of cumulative impacts in the APD approvals to the 2003 RMP.
We are not persuaded the district court abused its discretion in holding that Plaintiffs had failed to demonstrate a likelihood of success on the' merits of this argument. As' the district court pointed out, only 3,860 of the anticipated 9,942 new wells in the planning area were drilled in the twelve years between the issuance of the 2003 RMP and the court’s consideration of this issue in 2015. Thus, 'even with increased drilling in the Mancos Shale formation and the switch to horizontal drilling and multi-stage fracturing, the overall amount of drilling and related surface impacts are still well within the anticipated level.2 As for the possibly increased air quality impacts, the agency considered these impacts in its environmental assessments and concluded that the approved drilling activities would not cause a significant increase in emissions over the amount anticipated in the RMP or a violation of national air quality standards for any criteria pollutant. After reviewing the agency’s RMP and its assessment of site-specific and cumulative impacts in the APD approvals before us in this appeal, we are not persuaded that the district court abused its discretion in concluding that Plaintiffs have not shown a substantial likelihood of success on the merits of this argument.
We are similarly unpersuaded by Plaintiffs’ argument that they are likely to succeed- on the merits of their claims be*1284cause new horizontal drilling and multistage fracturing technologies will lead to environmental impacts qualitatively different from the impacts assessed in the 2003 RMP. Plaintiffs argue there are two main reasons why these new technologies will cause different types of impacts from the impacts anticipated in the RMP. First, these technologies allow operators to extract significant amounts of oil from the Mancos Shale, while the RMP mainly anticipated the extraction of gas from other formations in a different region of the San Juan Basin. Second, horizontal drilling and multi-stage fracturing “involve a number of.complexities not associated with conventional wells” that “could result in additional . environmental impacts that were not anticipated or analyzed when the agency analyzed the impacts of conventional drilling methods in the 2003 [RMP].” (Appellants’ Opening Br. at 21 (internal quotation marks omitted).)
■ Neither of these arguments is sufficiently developed or supported to sustain the conclusion that Plaintiffs are likely to succeed on the merits of their claim, given the deferential agency standard of review at issue in this case. Plaintiffs fail to present any argument or cite to any evidence as to how drilling in the Mancos Shale will cause different environmental impacts than drilling in other formations in the San Juan Basin—including the anticipated “Dakota, Mancos” drilling that the RFDS anticipated would account for more than forty percent of the new wells in the planning area over the next twenty years—or as to how additional oil wells will cause qualitatively different impacts from the smaller number of oil wells and larger number of gas wells anticipated in the RMP. Plaintiffs have likewise failed to present us with any argument or evidence to support their contention that horizontal drilling and multi-stage fracturing may give rise to different types—rather than just different levels— of environmental harms when compared to the traditional vertical drilling and hydraulic fracturing techniques that have historically been used in the San Juan Basin. Since the plaintiff in an environmental case has the burden of proving that the agency action was invalid, see Citizens’ Comm, to Save Our Canyons v. Krueger, 513 F.3d 1169, 1176 (10th Cir. 2008), Plaintiffs’ failure to support their arguments on this issue is fatal to their argument that the district court abused its discretion in denying a preliminary injunction due to the different types of environmental harms allegedly caused by horizontal drilling and multi-stage fracturing.
Finally, Plaintiffs argue that the BLM’s continued approval of new wells is imper-missibly limiting the BLM’s options as it amends the 2003 RMP, thus violating the regulations and satisfying the arbitrary and capricious standard of agency review. They also raise the related argument that the BLM’s decision to amend the 2003 RMP to account for increased development in. the Mancos Shale proves that the 2003 RMP is inadequate and thus .the BLM acted arbitrarily and capriciously by referring to the 2003 RMP in its environmental assessments in the contested APDs.
When an agency has not yet prepared an environmental impact statement for a proposed activity, it may not take an action relating to the proposed activity that would either have an adverse environmental impact or limit its choice of reasonable alternatives. See 40 C.F.R. § 1506.1. Plaintiffs argue the same reasoning must apply when an agency is amending its environmental impact statement to account for new developments affecting the planning area. However, Plaintiffs cite to no authority for their proposition that 40 C.F.R. § 1506.1 applies to interim development under an existing environmental impact statement and RMP. Moreover, we are not *1285persuaded that the BLM’s continued approval of wells under an existing RPM that allows for continued drilling in the planning area will impermissibly “limit the choice of reasonable alternatives” in the BLM’s revision of its RMP.3 We are also unpersuaded by Plaintiffs’ argument that the BLM’s decision to revise the RMP is evidence that the agency has acted arbitrarily and capriciously by referring to the 2003 RMP .in its environmental assessments of the contested APDs. The agency’s decision to improve its plan for managing federal lands in the San Juan Basin does not immediately invalidate the old plan or prevent the agency from referring to it. ■ ' ■ •
While Plaintiffs identify several areas in which they believe the BLM’s methodologies and conclusions fall short, their disagreements with the agency’s decision are insufficient to convince us that the district court abused its discretion in concluding they had not shown a substantial likelihood of success on their claim that the agency acted arbitrarily and capriciously in approving the APDs. We therefore affirm on this ground and do not address the parties’ arguments regarding the other three prerequisites for preliminary relief. We note that this decision does not foreclose■ an ultimate ruling in Plaintiffs’ favor. The record'before us in this interlocutory appeal is not the final administrative record that will be used to resolve the merits of this case, and additional evidence, better-developed arguments, or further consideration may lead to the conclusion that the agency actually failed to adequately evaluate 'the environmental impacts of its decisions to grant the APDs. Based on the record and arguments before us in this appeal, however, Plaintiffs have failed to persuade us the district court abused its discretion in concluding that they have not' shown a substantial likelihood of success under the deferential standard of review applicable to judicial review of agency actions. -
III.
We accordingly AFFIRM the district court’s denial of Plaintiffs’ request fór a preliminary injunction.
. Plaintiffs do not request preliminary relief for wells that have already been drilled. They assert that their request for a preliminary injunction involves a total of 142 APDs that have been approved but not yet drilled.
. The largest difference Plaintiffs identify- between horizontal and vertical drilling relates to the amount of water used in drilling. According to Plaintiffs' expert, hydraulically fracturing a horizontal well requires five to ten times more water than hydraulic fracturing of a vertical well. However, the BLM presented its own evidence that efforts áre being made to recycle flowback water and that operators are using nitrogen foam to reduce the amount of water needed. Moreover, the BLM’s field manager for the Farm-ington Office has asserted that the agency is not required to estimate the amount of water being used for drilling in the San Juan Basin because all water used in drilling and fracturing operations comes from permitted sources approved by the New Mexico State Engineer's Office. "Because the State of New Mexico has jurisdiction over water use, the State analyzes the impacts of potential groundwater depletion caused by these wells and has not determined that the current use will lead to depletion.” (Joint App. at 680.) Plaintiffs do not address or even acknowledge these assertions. Although we express some concern over the BLM’s apparent disavowal of any independent responsibility to consider whether hydraulic fracturing of horizontal wells may result in groundwater depletion in the San Juan Basin, it is Plaintiffs’ burden in this interlocutory appeal to convince us that the district court abused its discretion in denying extraordinary relief, and we will not make Plaintiffs’ arguments'for them. Given the agency's position, Plaintiffs’ bare assertion that horizontal wells require five to ten times as much water as vertical wells falls far short of meeting this burden.
. Notably, “[wjhen an RMP is revised after approval of a lease or an APD, those revisions may apply to existing leases and site-specific activities (such as operation of existing roads or wells).” (Joint App. at 666.) While it is conceivable that the BLM would decide to revise the RMP to reduce the number of wells it will approve in the Mancos Shale formation or call for the" reclamation of additional drilled mines, Plaintiffs ■ have failed to persuade us that the approval of a few more wells in an area that has been extensively mined for more than half a century will somehow limit the BLM’s'choice of alternatives for managing federal lands in the San Juan Basin.