OPINION

Rosaeker appeals the sentence imposed by the district court after he pleaded guilty to one count of using a communication device to facilitate a drug trafficking offense in violation of 21 U.S.C. § 843 (b). The district court sentenced him to the statutory maximum, 48 months. On appeal, Rosaeker argues that the district court erred by (1) basing the estimate on an unreliable police laboratory report, and (2) applying the preponderance of the evidence standard rather than the clear and convincing standard in making the drug quantity approximation. We vacate the sentence and remand because we conclude that the police lab report was unreliable. We affirm the district court’s use of the preponderance standard.

I. BACKGROUND

On February 10, 2001, Rosacker’s mother discovered what appeared to be a drug lab in a shed on her property that Rosaeker had been using and reported her discovery to the Yamhill County Sheriffs Office. Law enforcement officers from various agencies searched the shed and found twenty-six plastic bags labeled “Herba Ephedrae” and filled with one pound each of “Ma Huang” tea (“mahuang”) and ten empty bags of the same type. The officers also found a five-gallon plastic bucket containing mahuang, two Mason jars containing a three-layered liquid, a funnel containing a coffee filter with an oily brown sludge, a Mason jar with reddish stains, a propane tank, and a pressure cooker with a reddish stain.

The officers searched Rosacker’s pickup truck and found a 400 milliliter glass cooker, pH strips, a receipt for the purchase of toluene, denatured alcohol, muriatic acid, two containers of Red Devil lye, some clear plastic tubing, and a book entitled “Secrets of Methamphetamine Manufacture.” The officers conducted a walk-through of Rosacker’s residence and observed numerous large containers holding mahuang or mahuang-and-liquid mixtures, a twenty-quart cooker containing mahuang and lye being heated on the stove, containers of muriatic acid and denatured alcohol, Red Devil lye, a homemade strainer with tubing leading to a bathtub drain that was surrounded by dark brown stains, another strainer containing mahuang, two buckets connected by rubber tubing, funnels, and a spatula caked with mahuang.

Samples from the shed, the pickup, and Rosacker’s residence tested positive for ephedrine and pseudoephedrine. Rosaeker told the officers that he was attempting to extract ephedrine from the mahuang in order to make methamphetamine, but he was unsure how to create a “good batch” of the drug. Rosaeker pleaded guilty to the charge of using a telephone to facilitate the attempted manufacture of a controlled substance, in violation of 21 U.S.C. § 843 (b).

The presentence report (“PSR”) stated that Rosaeker “may not have been able to [extract] the precursor chemical, pseu-doephedrine, from the Ma Huang tea.” However, based on a one-page report prepared by the Oregon State Police Forensic Laboratory (“forensic lab report”), the PSR also stated that Rosaeker could have produced 80 grams of methamphetamine and recommended a base offense level of 32. Rosacker objected to these portions of the PSR and submitted an expert’s report that criticized the forensic lab’s assumptions, methods and conclusions. Rosacker’s expert stated in the report that Ro-sacker had neither the knowledge nor the equipment necessary to extract ephedrine from mahuang. The expert also stated that “over the course of fifteen years, I have never seen a case where the methamphetamine was made from ephedrine or pseudoephedrine extracted from Ma-Huang tea.”

The PSR recommended a sentence of 48 months, the maximum allowed by the statute, because the 80-gram drug quantity estimate corresponded to a guideline range of 108-135 months. Rosacker argued that he could not have produced any methamphetamine from the mahuang, and therefore, the applicable sentencing range was 10-16 months.

Applying the preponderance standard, the district court accepted the Government’s argument that the applicable guideline range exceeded the 48-month statutory maximum even if Rosacker could have produced only 5 grams of methamphetamine. The court then orally found that “Mr. Rosacker could have produced at least five grams of methamphetamine,” and sentenced him to 48 months.

Rosacker challenges both the district court’s estimate of the quantity of methamphetamine he could have produced and the evidentiary standard the district court applied. This court reviews a district court’s interpretation and application of the guidelines de novo. It also reviews de novo whether a district court’s method of approximation of the relevant drug quantity conforms to the guidelines. The district court’s factual findings are reviewed for clear error.

II. RELIABILITY OF THE POLICE FORENSIC LABORATORY REPORT

The state police forensic laboratory report was not sufficiently reliable to support the district court’s drug quantity approximation. Drug quantity calculations in the forensic lab report rested on four unsupported assumptions: (A) that 1% of the mahuang was extractable ephedrine or pseudoephedrine, (B) that the ephedrine and pseudoephedrine could be extracted from the mahuang, (C) that Rosacker was personally capable of extracting ephedrine or pseudoephedrine and converting it to methamphetamine, and (D) that 100% of the ephedrine and pseudoephedrine present in the mahuang could be extracted.

A. Assumption of 1% Precursor Content

“[T]he information which supports an approximation must possess ‘sufficient indicia of reliability to support its probable accuracy.’ ” In other words, a drug approximation must have a “reliable eviden-tiary basis.” Accordingly, a court “ ‘may not ... adopt conclusory statements [from the PSR] unsupported by facts or the Guidelines.’ ”

The forensic lab report set forth a “conservative estimate” that 1% of the mahuang could be extracted as ephedrine or pseudoephedrine (collectively, “precursor chemicals”). However, the report supplied no data, rationale, discussion, or any other evidentiary basis in support of the 1% figure. In addition, Rosacker’s expert report and the forensic lab report agreed that the precursor chemical content of mahuang varies. Rosacker’s expert report — • to which the Government did not object and which the district court entered into the record — also stated that the content depends on the particular species of mahuang and ranges from zero to 1%.

The report’s characterization of the 1% estimate as “conservative” does not substitute the lack of a “reliable evidentiary basis” for the estimate. “Any estimate— including a conservative one — must be supported by reliable information.” Moreover, the characterization of 1% as a “conservative” estimate appears questionable. Rosacker’s expert report cited ephedrine/pseudoephedrine content figures ranging from zero to 1%, but none higher than 1%. The Government did not, and does not, dispute this range. It can hardly be said then, that the 1% maximum extraction rate is a “conservative” estimate. Thus, not only did the Government’s forensic lab report fail to supply any evidentiary basis for the 1% figure, evidence on the record provided affirmative reasons for doubting the reliability of the 1% estimate. Accordingly, the forensic lab report’s 1% content estimate was a conclusory statement unsupported by facts or the guidelines.

B. Assumption that Precursor Chemicals can be Extracted From Ma-huang

Rosacker could have produced methamphetamine only if it was possible to extract precursor chemicals from the mahuang. The forensic lab report’s calculations implicitly assumed that extraction was possible. However, the Government offered no evidence in support of this assumption. The report does not describe, explain, refer to, or even offer conjecture regarding any process by which any precursor chemicals could have been extracted from the mahuang. In addition, there is no evidence that Rosacker ever actually succeeded in extracting precursor chemicals in a form usable for methamphetamine production. Accordingly, the record upon which the district court based its drug approximation contains no evidence that precursor chemicals could be extracted from the mahuang.

Evidence on the record suggests the contrary — that it may not be possible to use mahuang as a source of precursor chemicals for the production of methamphetamine. Rosacker’s expert wrote that he had analyzed evidence from numerous methamphetamine labs for fifteen years but had never seen a case in which precursor chemicals had been extracted from ma-huang.

Under these circumstances, the forensic lab report’s implicit assumption that precursor chemicals could be extracted from the mahuang lacks support. The assumption is another conclusory statement unsupported by facts or the guidelines. Accordingly, the report did not provide a “reliable evidentiary basis” for approximating the amount of precursor chemicals — and thus the amount of methamphetamine — that could be produced from.the mahuang.

C. Assumption that Rosacker Could Extract the Precursor

“The capability of a drug operation is a factual issue reviewed for clear error.” In estimating drug quantity, “[t]he district court must conclude that the defendant is more likely than not actually responsible for a quantity greater than or equal to the quantity for which the defendant is being held responsible.” The guidelines specifically permit the district court to estimate drug quantities. However, the district court “shall” not consider amounts that a defendant proves he was not capable of producing.

Other courts have required district courts to approximate drug quantities by evaluating the actual ability of the defendant’s laboratory to manufacture the drug. For example, in United States v. Eschman, the court reversed a district court sentence because it lacked an evidentiary foundation with which to estimate the drug quantity. Acknowledging that U.S.S.G. § 2D 1.1 allows for approximations of drug quantity, the Seventh Circuit required the district court to base its estimate “on reliable evidence.” The court held that “district courts cannot quantify yield figures without regard for a particular defendant’s capabilities when viewed in light of the drug laboratory involved.” The court remanded to the district court to make “a more precise inquiry” into the defendant’s actual ability and the capabilities of his lab.

Actual ability of the lab in question has been a factor in assessing a defendant’s capability in the Eighth Circuit as well. In Anderson, the court said that “the relevant inquiry [wa]s not what a theoretical maximum yield would be, or even what an average methamphetamine cook would produce, but what appellants themselves could produce.” Expert testimony established that Anderson’s co-conspirator had a methamphetamine recipe capable of producing the drug. Additionally, trace amounts of methamphetamine were located at Anderson’s lab. Holding that this evidence established the lab’s capabilities despite the fact that some necessary ingredients were missing from the lab, the court stated, “an estimate of a laboratory’s capability based on the quantity of precursor chemicals seized need not be limited to the yield available from the least abundant precursor chemical.”

The Tenth Circuit upheld a sentence based on approximate drug manufacturing potential, despite missing ingredients and processing equipment as well. In Havens, the defendant pleaded guilty to attempting to manufacture methamphetamine. Finding that the district court based its sentence on sufficient testimony at the sentencing hearing, the court said that “[t]he factual question is what each specific defendant could have actually produced, not the theoretical maximum amount produceable [sic] from the chemicals involved.” The missing ingredients were accessible enough that the court could approximate the drug manufacturing capabilities as if they were present.

Rosacker’s circumstances differ from Havens and Anderson because the forensic lab report and PSR contained no evidence that suggested Rosacker had the ability to extract precursor chemicals from the mahuang or to convert the precursor chemicals to methamphetamine. The PSR even stated that Rosacker “may not have been able” to extract precursor chemicals from the mahuang. Rosacker’s lab equipment — mainly plastic buckets, tubing and an assortment of kitchen-ware — was primitive and unsophisticated.

The defense expert opined that Ro-sacker’s lab was not capable of producing methamphetamine. To convert the precursor chemical extract into methamphetamine, Rosacker would have needed “further chemical ingredients, advanced laboratory equipment and glassware, (things that are highly monitored and controlled by the DEA), such as: hy-driodic acid, round bottom flasks with condenser and heating mantles, trichloro-trifluorethane, and sodium sulfate.” The Government did not refute Rosacker’s expert testimony.

In sum, the record contains no affirmative evidence that Rosacker could extract precursor chemicals or convert them to methamphetamine; but it does contain evidence suggesting that he lacked the requisite lab capability. Accordingly, neither the facts nor the guidelines support the forensic lab report’s assumption that Ro-sacker could extract precursor chemicals and convert them to methamphetamine.

D. Assumption of 100% Extraction Rate

The forensic lab report assumed that 1% of the mahuang was extractable precursor chemicals and implicitly assumed that Ro-sacker could have extracted the full 1%. Given that the report provided no basis for Rosacker’s ability to extract any chemicals, the idea that he could extract 100% of the precursor chemicals is entirely unsupported.

Thus, the PSR and the forensic lab report contain no reliable evidentiary basis for any of the pivotal assumptions in the drug quantity approximation. Accordingly, the district court erred in adopting the PSR’s conclusion (based on the forensic lab report) that Rosacker could have extracted precursor chemicals and converted them to methamphetamine.

III. EVIDENTIARY STANDARD

Rosacker contends that the district court erred in applying the preponderance of the evidence standard rather than the clear and convincing standard in making the drug quantity approximation. Rosacker argues that the clear and convincing standard should be applied to his upward departure from the sentencing guidelines, as they have been applied in other “extremely disproportionate” sentences. In Mezas de Jesus, the court applied the clear and convincing standard, rather than the preponderance standard, because the court considered other crimes for which the defendant was not charged, raising the defendant seven levels in the guidelines. We do not agree that the “extremely disproportionate” sentencing enhancement standard applies with equal force to the drug quantity approximations.

United States v. Harrison-Philpot forecloses Rosacker’s application of Mezas de Jesus. In Harrison-Philpot, the court observed that the “distinction between a quantity determination and [the] uncharged criminal conduct” that supports a sentence enhancement places drug quantity approximations “on a fundamentally different plane” than sentencing enhancements. Accordingly, drug quantity approximations “do[ ] not provide the legal basis for the due process concerns” that exist in the context of enhancing a sentence on the basis of uncharged conduct. Thus, the district court did not err in applying the preponderance standard.

IV. CONCLUSION

The district court correctly applied the preponderance of the evidence standard. However, the forensic lab report adopted by the district court in applying Rosacker’s sentence was fundamentally unreliable. Accordingly, we vacate the sentence imposed and remand to the district court for resentencing, and we affirm the evidentia-ry standard that the district court applied.

VACATED and REMANDED in part and AFFIRMED in part.

. Rosaeker also argues that the Government failed to prove that he had the knowledge and capability to produce methamphetamine. However, this argument is part of the larger question of the reliability of the forensic lab report and is addressed in that section. He further argues that even if he could produce only 5 grams of methamphetamine, his sentencing guideline range exceeded the statutory maximum. Because we vacate the sentence and remand to the district court on another issue, we do not reach this issue. . See 21 U.S.C. § 843 (d). . See u.S. Sentencing Guideline § 2Dl.l(c) (2001)(U.S.S.G.). Rosacker’s criminal history category was III, which would warrant a 10-16 month sentence if there was no drug quantity finding. . The court’s subsequent written order listed 50 grams, not 5 grams, as the amount of methamphetamine that Rosacker could have produced. Because we vacate and remand to district court on other grounds, we need not address this disparity. . United. States v. Culps, 300 F.3d 1069, 1076 (9th Cir.2002). . Id. . Id. . "Where the amount of drugs seized does not reflect the scale of the offense, the district court may approximate the quantity of the drugs.” Id.; see also U.S.S.G. § 2Dl.l(c) cmt. n. 12. . Culps, 300 F.3d at 1076 (quoting United States v. Walton, 908 F.2d 1289, 1302 (6th Cir.1990) (quoting U.S.S.G. § 6A1.3(a))). . Id. at 1077. . Id. at 1078 (quoting United States v. Navarro, 979 F.2d 786, 789 (9th Cir.1992)). . Id. at 1079 n. 5. . The Government does not argue that Ro-sacker could have used the raw mahuang as precursor chemicals for methamphetamine conversion. .The Government has the burden of proving the facts that support the drug quantity approximation. See Culps, 300 F.3d at 1076 (stating that "the government is required to prove the approximate quantity by a preponderance of the evidence”) (internal quotation marks and citation omitted). . United States v. Bertrand, 926 F.2d 838, 846 (9th Cir.1991). . Culps, 300 F.3d at 1076 (emphasis added) (internal quotation marks and citations omitted). . U.S.S.G. § 2D 1.1(c) cmt. n. 12 (“Where there is no drug seizure or the amount seized does not reflect the scale of the offense, the court shall approximate the quantity of the controlled substance. In making this determination, the court may consider ... the size or capability of any laboratory involved.") (emphasis added). . Id. (“If, however, the defendant establishes that he ... was not reasonably capable of providing! ] the agreed-upon quantity of the controlled substance, the court shall exclude from the offense level determination the amount of controlled substance that the defendant establishes that he ... was not reasonably capable of providing.”) (emphasis added). While comment note 12 is an example of an agreement to sell a specified quantity of drugs, it follows that the district court should consider the defendant's ability to manufacture a drug when establishing the severity of the offense. . 227 F.3d 886 (7th Cir.2000). . Id. at 890 . . Id. . Id. at 891 . . See United States v. Anderson, 236 F.3d 427, 430 (8th Cir.2001) (per curiam). . Id. at 430 . . Id. . Id. . Id. at 430 n. 5. . United States v. Havens, 910 F.2d 703, 705 (10th Cir.1990). . Id. at 704 . . Id. at 706 (emphasis added). . See id. at 704 ; see also United States v. Barnett, 989 F.2d 546, 553 (1st Cir.1993) (stating that "the quantity of essential precursor chemicals seized, like the capacity of the laboratory and the evidence relating to the overall scheme, is but one among several circumstantial factors appropriately considered in approximating drug quantities for sentencing purposes”) (citation omitted). . Rosacker has prior methamphetamine charges and convictions but none for manufacturing or attempting to manufacture methamphetamine or any other drug. . The Government argues that the district court was entitled to assume the presence of precursor chemicals, citing United States v. Basinger, 60 F.3d 1400 (9th Cir.1995). In Basinger , however, the drug quantity approximation was based on empty ephedrine containers. Id. at 1406 . In other words, there was evidence supporting an inference that the defendant had possessed ephedrine precursor chemicals. In contrast, the Government produced no evidence in this case supporting an inference that Rosacker ever possessed or successfully extracted precursor chemicals. This case also differs from Basinger in that, under the Government's theory, Rosacker had to extract — not purchase — precursor chemicals in order to manufacture methamphetamine. The Government, therefore, had the burden of proving that Rosacker could, or did, extract precursor chemicals from the ma-huang. The Government failed to produce any such evidence. Accordingly, acceptance of the assumption that Rosacker could have, or did, extract precursor chemicals would im-permissibly operate in this case to relieve the Government of its burden of proving the facts that support a drug approximation. . See Culps, 300 F.3d at 1078 . . The Government argues that Rosacker waived any challenge to the forensic lab report and to the district court's finding that Rosacker could have produced at least 5 grams of methamphetamine. We do not agree. Rosacker expressly objected to the portions of the PSR that incorporated the forensic lab report's calculations, and the sentencing hearing focused largely on the reliability of the forensic lab report. In addition, the district court predicated its finding of 5 grams on the forensic lab report's assumption that ephedrine could be extracted from ma-huang. Rosacker directly challenged this assumption at the hearing and the district court squarely recognized it as one of the sentencing issues. Under these circumstances, Ro-sacker did not waive any objection to the forensic lab report or the 5-gram finding. See Bothell v. Phase Metrics, Inc., 299 F.3d 1120, 1130 (9th Cir.2002) (finding no waiver when the issue was "clearly raised and argued” and the district court "considered it on its merits”). . United States v. Mezas de Jesus, 217 F.3d 638, 639-40 (9th Cir.2000); see also id. at 641 n. 7 & 642 (stating that "when a sentencing factor has an extremely disproportionate ef- feet on the sentence relative to the offense of conviction ... a higher standard of proof may be required”) (internal quotation marks and citation omitted). . Id. at 645 . . 978 F.2d 1520 (9th Cir.1992). . Id. at 1523 . . Id. . Id.