Defendants-Appellants Kevin Sims, Lut Muhammad, and Okeiba Sadio appeal from their judgments of conviction entered on January 11, 2012. Sims, Muhammad, and Sadio were indicted on December 2, 2009 on various narcotics charges related to a cocaine base — or “crack” cocaine — conspiracy led by twin Bronx residents William and Wilson Pena. Sims pleaded guilty to one count of conspiracy to possess with intent to distribute, and to distribute, 5 grams or more of cocaine base in violation of 21 U.S.C. §§ 841(a)(1), 841(b)(1)(C), and 846, and the District Court imposed a sentence principally of 120 months’ imprisonment and five years of supervised release. Muhammad pleaded guilty to eleven counts, including conspiracy to possess with intent to distribute, and to distribute, 50 grams or more of cocaine base. He was convicted of violating 21 U.S.C. §§ 841(a)(1), 841(b)(1)(B), 846, and 851 and sentenced to concurrent terms of 240 months’ imprisonment, followed by eight years of supervised release. After trial, a jury convicted Sadio of two counts: conspiracy to possess with intent to distribute, and to distribute, 50 grams or more of cocaine base, in violation of 21 U.S.C. § 846, and possession with intent to distribute 50 grams or more of cocaine base, in violation of 21 U.S.C. §§ 841(a)(1) and 841(b)(1)(B). The District Court sentenced Sadio to concurrent terms of 240 months’ imprisonment, followed by eight years of supervised release. We assume the parties’ familiarity with the facts and record of the prior proceedings, to which we refer only as necessary to explain our decision to affirm.
1.
Kevin Sims
Sims purchased crack cocaine from other members of the conspiracy, including Muhammad and co-defendant Emmanuel Tyson, and sold some of it to support his drug addiction. Sims appeals his sentence on two grounds. First, he argues that the District Court erroneously attributed more than 112 grams of crack cocaine to him. Second, Sims argues that the District Court unreasonably failed to consider his mental illness and drug dependency as mitigating factors. We address each argument in turn.
a.
Drug Quantity
“A Sentencing Guidelines calculation must begin with an identification of the defendant’s relevant conduct, which in the case of a drug ... offense includes the quantity of drugs controlled by the defendant. ...”
United States v. Jones,
531 F.3d 163, 174-75 (2d Cir.2008). “The quantity of drugs attributable to a defendant is a question of fact,” which we review for clear error.
Id.
at 175-76. Where drugs were not seized from the defendant, “the court shall approximate the quantity of the controlled substance,” U.S.S.G. § 2D1.1 app. n. 5, based on a preponderance of the evidence,
id.
at § 6A1.3. “To sustain quantity-based enhancements for relevant conduct, the court must base its findings on ‘specific evidence’ ” —which can be circumstantial — such as “drug records, admissions or live testimony.”
United States v. Archer,
671 F.3d 149, 162 (2d Cir.2011).
The District Court held an evidentiary hearing to determine the quantity of cocaine base attributable to Sims. At that hearing, Emmanuel Tyson testified that he sold one to two grams of crack cocaine to Sims two to three times per week, from July 2009 to November 2009. The District Court found Tyson credible and, based on that range of quantities and frequencies, calculated that in a 20-week period, Tyson sold Sims 75 to 80 grams of crack cocaine. Wiretap evidence revealed that between September 24, 2009, and November 3, 2009, Sims purchased crack cocaine from Lut Muhammad two or three times a week
in quantities ranging from two to five grams. The District Court reasoned that Sims purchased an estimated 70 to 74 grams of cocaine base from Muhammad. The court then estimated that Sims bought a total of 145 to 154 grams of crack cocaine from Tyson and Muhammad. As an offender responsible for between 112 and 196 grams of cocaine base, Sims’s base offense level was 28, yielding a Guidelines range of 120 to 150 months’ imprisonment.
See
U.S.S.G. § 2Dl.l(c)(6).
Sims asserts that the District Court erred by relying on Tyson’s testimony at the hearing, as Tyson previously stated that Sims purchased drugs from him only eight or nine times, and his statements at the hearing with regard to the number of sales were somewhat contradictory. Sims also argues that the District Court should have discounted Tyson’s testimony because Tyson was a drug abuser with an extensive criminal record. “[Cjredibility determinations are the province of the trial judges, and should not be overruled on appeal unless clearly erroneous.”
United States v. Yousef,
827 F.3d 56, 124 (2d Cir.2003) (quotation marks omitted). At the hearing, Tyson clarified that when he previously stated that he sold crack to Sims eight or nine times, he may have believed the question referred to a single month. Tyson otherwise consistently stated that he dealt with Sims approximately two or three times per week. It was within the court’s province to determine that Tyson’s estimate of the number of sales per week more accurately represented the number of times Tyson sold narcotics to Sims. Our own review of the transcript of Tyson’s testimony supports the District Court’s conclusion that Tyson was a witness who “knew his business well and was able to explain how it operated clearly.” Accordingly, we reject Sims’s argument that the District Court erred by crediting and relying on Tyson’s testimony to determine the quantity of cocaine base attributable to Sims.
Similarly we reject Sims’s argument that the District Court’s estimate of drug quantity was not reliable and conservative, as it should have accounted for the possibility that Tyson sold to Sims only eight or nine times. “[Gjiven the wide latitude of the district court to make credibility determinations, the court is not restricted to accepting the low end of a quantity range estimated by a witness.”
United States v. Blount,
291 F.3d 201, 215 (2d Cir.2002) (citation omitted). With that principle in mind, we conclude that the District Court was free to credit only Tyson’s testimony at the hearing that he sold to Sims two to three times a week. We cannot say the District Court’s estimation rested on pure speculation or that it clearly erred in its determination of drug quantity.
b.
Mitigating Factors
We also reject Sims’s argument that the District Court imposed a procedurally and substantively unreasonable sentence by failing to decrease his sentence based on his history of mental illness and drug abuse.
See United States v. Cavera,
550 F.3d 180, 200 (2d Cir.2008) (en banc). Sims argued to the District Court that his mental health problems, including his drug dependence, were relevant personal characteristics for the purpose of 18 U.S.C. § 3553(a)(1). He asked the District Court to depart downward from the applicable Guidelines range because he personally used much of the crack that he purchased instead of distributing it, such that the Guidelines range overstated the seriousness of his offense conduct. At sentencing, the District Court said that it had considered each of the § 3553(a) factors along with Sims’s arguments. However, after also considering a May 2011 report
from the Bureau of Prisons, which stated that Sims was malingering, along with the transcripts of recorded conversations and Sims’s in-court behavior, the District Court declined to find that Sims’s mental health problems significantly contributed to his offense. We cannot say that this decision, based on the evidence presented to the District Court and its own first-hand observation, was error. Further, it was within the District Court’s wide discretion to determine that specific deterrence was the most important factor in sentencing Sims and that a sentence of ten years was sufficient but not greater than necessary to achieve that objective. Accordingly, we affirm Sims’s sentence.
2.
Lut Muhammad
Muhammad, the original target of the Government’s 2009 investigation, purchased crack cocaine in the Bronx and distributed it in and around Stamford, Connecticut. He challenges his sentence on appeal, arguing that the District Court erred by sentencing him based on drug quantities that were neither voluntarily pleaded nor proved to a jury. Specifically, Muhammad claims that the court erred when it determined that he purchased at least 2.8 kilograms of cocaine base during the course of the conspiracy. For the following reasons, we reject Muhammad’s arguments and affirm his sentence.
a.
The Guilty Plea
On September 7, 2010, just over a month after the Fair Sentencing Act lowered the penalties for crack cocaine offenses,
see
Fair Sentencing Act of 2010, Pub.L. No. 111-220, 124 Stat. 2872, Muhammad entered into a plea agreement with the Government and pleaded guilty to counts one through eleven of the second superseding indictment. By pleading guilty to count one of the indictment, Muhammad acknowledged that he conspired to distribute at least 50 grams of cocaine base. The plea agreement stated that this conduct violated 21 U.S.C. §§ 841(a)(1), 841(b)(1)(A) and 846. It therefore provided that Muhammad, who had a prior felony drug conviction, faced a mandatory minimum sentence of 20 years under § 841(b)(1)(A). The Government now concedes that this was incorrect. The higher drug quantity thresholds implemented by the Fair Sentencing Act apply to offenders, like Muhammad, whose criminal conduct occurred prior to the Act but who were sentenced after the Act took effect.
See Dorsey v. United States,
— U.S. -, 132 S.Ct. 2321, 2331, 183 L.Ed.2d 250 (2012). Therefore, to be subject to the 20-year mandatory minimum in § 841(b)(1)(A) under the Act, Muhammad would need to have conspired to distribute at least 280 grams of crack cocaine.
See
21 U.S.C. § 841(b)(l)(A)(iii). If he possessed only 50 grams — more than 28 grams but less than 280 grams — he would be subject to the 10-year mandatory minimum contained in § 841(b)(1)(B). The District Court acknowledged at sentencing that § 841(b)(1)(B) was the applicable penalty provision and that Muhammad was subject to that ten-year mandatory minimum sentence. Nevertheless, it sentenced Muhammad to 20 years’ imprisonment based on the Guidelines and the factors listed in § 3553(a). As counsel acknowledged at oral argument, Muhammad never sought to withdraw or otherwise challenge the sufficiency of his guilty plea before the District , Court, and our review is therefore for plain error.
See United States v. Garcia,
587 F.3d 509, 515 (2d Cir.2009).
We reject Muhammad’s argument that the District Court erred under
United States v. Gonzalez,
420 F.3d 111 (2d Cir.2005). In that case, Gonzalez did not admit to, and in fact disputed, the drug quantity element of § 841(b)(1)(A), and that
element was never proven to a jury.
Gonzalez,
420 F.3d at 115. We therefore held that his plea “at best supports conviction on a lesser, unquantified drug charge, whose sentencing range is prescribed by § 841(b)(1)(C).”
Id.
Here, by contrast, Muhammad admitted that his conduct involved at least 50 grams of crack cocaine in his plea allocution and in his plea agreement, both of which supported a conviction for violating § 841(b)(1)(B). We find no support for Muhammad’s argument that the more severe penalties in place before the Fair Sentencing Act cast doubt on the voluntariness of his plea. We therefore find no error in his sentence.
b.
The Sentencing
In connection with Muhammad’s sentencing the District Court held a quantity hearing on January 4, 2012. Based on all of the evidence presented at the hearing, the District Court determined that 3,829 grams of crack cocaine were attributable to Muhammad and sentenced him accordingly. Muhammad argues that the District Court clearly erred by finding that he purchased over 2.8 kilograms of crack cocaine during the conspiracy. For the following reasons, we disagree.
We first reject Muhammad’s argument that William Pena’s testimony was not “specific evidence” supporting a finding that Muhammad purchased 1,040 grams of crack cocaine between May 15 and August 28, 2009.
United States v. Shonubi,
998 F.2d 84, 89 (2d Cir.1993). Pena testified that, starting in the spring of 2009, Muhammad bought crack cocaine from him every four to five days. He also testified that in May and June 2009, Muhammad bought 40 or 50 grams of crack at each purchase and that this number increased gradually until shortly before Pena’s arrest, when Muhammad was buying 140 or 150 grams at each purchase. Based on this testimony, the District Court found that Muhammad purchased 1,040 grams of crack cocaine between May 15 and August 28, 2009.
Pena’s description of Muhammad’s course of conduct was sufficiently specific to support the District Court’s finding. This case is not like
Shonubi
or
United States v. Martinez,
133 Fed.Appx. 762 (2d Cir.2005) (summary order), where a district court extrapolated the drug quantities involved in a few transactions to other incidents in which the quantity of drugs involved was unknown. Here, Pena testified that Muhammad purchased 40 or 50 grams each time he visited, every four or five days; the District Court did not need to speculate in order to estimate the total amount of drugs that Muhammad purchased from Pena. In light of Pena’s testimony, we discern no error in the District Court’s finding that Muhammad purchased 1,040 grams over a 106-day period.
We also reject Muhammad’s argument that the District Court erred by attributing 124 grams of crack cocaine to Muhammad based on the $4,300 cash he possessed in his bedroom at the time of his arrest and the price he paid for crack cocaine. “[A] district court may equate seized currency to a quantity of drugs, at least when a preponderance of the evidence indicates that the currency was used to purchase drugs.”
Jones,
531 F.3d at 176. Although it is possible that, as Muhammad argues, this money was obtained from legitimate work, the District Court did not clearly err in finding, by a preponderance of the evidence, that the most likely explanation for this large amount of cash was illegitimate.
3.
Okeiba Sadio
Sadio challenges his conviction on two grounds. First, he argues that the affidavit in support of the search warrant for his home failed to establish a nexus between the cocaine base he purchased in the Bronx and his Connecticut residence. Based on that argument Sadio asserts that the evidence found in the residence should have been suppressed. Second, he argues that the evidence presented at trial was insufficient to establish beyond a reasonable doubt that he intended to distribute crack cocaine and that he participated in a crack cocaine conspiracy. For the following reasons, we reject both arguments.
a.
The Search Warrant
“To establish probable cause to search a residence, two factual showings are necessary — first, that a crime was committed, and second, that there is probable cause to believe that evidence of such crime is located at the residence.”
United States v. Travisano,
724 F.2d 341, 345 (2d Cir.1983). To find probable cause, we require only “that there be a fair probability that the premises will yield the objects specified in the search warrant.”
Id.
at 346. We give substantial deference to a magistrate’s finding of probable cause.
Id.
at 345.
The affidavit in support of the search warrant established probable cause to believe that evidence of Sadio’s drug crimes would be found at his Connecticut residence. The affidavit described events on three separate days. On October 28, 2009, (1) an intercepted telephone call indicated that a subject wanted to buy drugs from the Pena brothers in the Bronx, (2) an officer observed Wilson Pena and Sadio engage in a transaction in Sadio’s car, and (3) another officer saw Sadio arrive home in Stamford, Connecticut 55 minutes later. On November 2, 2009, law enforcement intercepted a call wherein Sadio arranged to buy 80 grams of crack cocaine from Wilson Pena. Officers observed Pena get in and out of Sadio’s car in the Bronx, and another officer observed Sadio arrive home approximately 45 minutes later. Finally, on November 27, 2009, law enforcement intercepted a call in which Sadio arranged to purchase 135 grams of crack cocaine and five grams of powder cocaine, an officer observed Sadio enter and exit William Pena’s residence in the Bronx, and another officer saw Sadio arrive home and enter his Connecticut residence 38 minutes later. Sadio argues that these facts fail to establish probable cause because the officers did not actually follow Sadio home, so they could not have discounted the possibility that he deposited the drugs somewhere along the way. He notes that the affidavit mentions a woman in the car with Sadio in the Bronx but does not mention her arriving at his house, suggesting that she may have left and taken the drugs with her. He also points out that the officers did not aver that, based on their experience, narcotics dealers typically keep drugs in their homes.
We conclude that the facts averred in the search warrant affidavit established a “fair probability” that a search of Sadio’s home would yield crack cocaine. The short time between Sadio’s transactions in the Bronx and his arrival at his Stamford home provides a “reasonable inference” of a nexus between the drug purchases and his residence “based on common sense.”
United States v. Singh,
390 F.3d 168, 182 (2d Cir.2004) (quotation marks omitted). Although the magistrate judge who issued the warrant would have had more probable cause to believe Sadio kept drugs in his house if the officers had followed him home, this was not required under the
circumstances. The search of Sadio’s home did not violate his Fourth Amendment rights.
Even if the affidavit in support of the warrant had not established probable cause, the District Court would not have erred in declining to suppress the evidence because the officers who obtained the warrant acted “with objective good faith.”
United States v. Leon,
468 U.S. 897, 920, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984). For the same reasons we conclude that the affidavit provided probable cause, we conclude that the warrant application was not “so lacking in indicia of probable cause as to render reliance upon it unreasonable.”
United States v. Moore,
968 F.2d 216, 222 (2d Cir.1992). Therefore, the District Court properly admitted the evidence found in Sadio’s Connecticut residence.
b.
Sufficiency of the Evidence
“[A] defendant challenging the sufficiency of the evidence that led to his conviction at trial bears a heavy burden, as the standard of review is exceedingly deferential.”
United States v. Coplan,
703 F.3d 46, 62 (2d Cir.2012) (citation and quotation marks omitted). “[T]he relevant question is whether, after viewing the evidence in the light most favorable to the prosecution,
any
rational trier of fact could have found the essential elements of the crime beyond a reasonable doub.t.”
Jackson v. Virginia,
443 U.S. 307, 319, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979). Under this deferential standard, the evidence supported Sa-dio’s conviction for possession with intent to distribute cocaine, as well as his conspiracy conviction.
Although no witness at trial saw Sadio distribute drugs and the Government did not make any controlled buys from Sadio, the Government presented enough evidence that Sadio intended to sell the drugs that he purchased. First, law enforcement officers found 93.7 grams of crack cocaine in his bedroom, and an expert with the Drug Enforcement Administration testified that a “use quantity” of crack cocaine is approximately 0.1 grams. Second, recorded conversations between Sadio and the Pena brothers revealed that he frequently purchased large quantities of crack cocaine, including 50 grams on October 30, 2009, 80 grams on November 2,143 grams of crack cocaine and 7 grams of powder cocaine on November 10, and 135 grams of crack cocaine and 5 grams of powder cocaine on November 27. Third, the Pena brothers testified that Sadio sometimes complained about the quality of the drugs by telling them, “the people was saying that it was no good,” “ ‘they’re not liking it,’ ‘they’re not feeling it,’ ” or, if the drugs were good, “‘[tjhey’re lovin’ me.’” Fourth, during the search of Sadio’s bedroom, officers found $2,321 cash in his clothing pockets, even though he had reported a total of $800 in income from autumn of 2006 to autumn of 2010. Fifth, the search revealed six cell phones and approximately a dozen SIM cards. Finally, the search of Sadio’s room failed to reveal equipment typically associated with crack cocaine use, such as a pipe. Cumulatively, this evidence would permit a rational juror to find, beyond a reasonable doubt, that Sadio intended to distribute the drugs that he purchased.
The same evidence also would permit a rational juror to convict Sadio on the conspiracy charge. A defendant may be convicted of a conspiracy charge if he “knew of the existence of the scheme alleged in the indictment and knowingly joined and participated in it.”
United States v. Gore,
154 F.3d 34, 40 (2d Cir.1998) (quotation marks omitted). “If there be knowledge by the individual defendant that he is a participant in a general plan designed to place narcotics in the hands of ultimate
users, the courts have held that such persons may be deemed to be regarded as accredited members of the conspiracy.”
United States v. Rich,
262 F.2d 415, 418 (2d Cir.1959). However, even though a drug buyer and a drug seller necessarily agree to transfer narcotics illegally, “the objective to transfer the drugs from the seller to the buyer cannot serve as the basis for a charge of conspiracy to transfer drugs.”
United States v. Parker,
554 F.3d 230, 234 (2d Cir.2009). This buyer-seller exception is limited, as it “does not protect either the seller or buyer from a charge they conspired together to transfer drugs if the evidence supports a finding that they shared a conspiratorial purpose to advance other transfers, whether by the seller or by the buyer.”
Id.
at 235. Sadio does not challenge the existence of a conspiracy led by the Pena brothers but instead argues that he was a mere buyer and not a member of that conspiracy.
The jury had a sufficient basis to conclude beyond a reasonable doubt that Sadio and the Pena brothers shared the purpose of advancing Sadio’s sales to his own customers. As noted, the Penas testified that Sadio referred to third parties complaining about or complimenting his product, and the Penas assumed these third parties were his customers. Although “mere awareness on the part of the seller that the buyer intends to resell the drugs is not sufficient to show that the seller and the buyer share a conspiratorial intent to further the buyer’s resale,”
id.,
a seller whose business depends on selling wholesale quantities “may well realize that his buyers’ ability to buy and pay for substantial amounts of drugs, and hence, his profit, will depend on the buyers’ ability to resell.”
Id.
at 236. William Pena testified that he did not sell to people who “were just buying for use” and instead trafficked in quantities of no less than 5 grams. A jury could therefore infer that his business depended on selling wholesale quantities to other dealers, and he and his brother knew that Sadio was one such dealer.
“In such case, the liability of buyer and seller for having conspired together to transfer drugs ... depend[s] ... on a further showing of the seller’s interest, shared with the buyer, in the success of the buyer’s resale.”
Parker,
554 F.3d at 236. To prove that interest, the Government also presented evidence that the Pe-nas supported Sadio’s dealing by, for example, immediately replacing 135 grams of crack when he complained that it was not good. William Pena testified that it was important to him to make Sadio happy because Sadio “was bringing [him] money, a lot of money.” Both William Pena’s testimony and the evidence of specific sales to Sadio suggest that the Penas’ business relied on the regular sale of crack cocaine in bulk, which could only be accomplished if their buyers were selling their purchases.
In sum, the evidence was sufficient to permit a rational jury to conclude beyond a reasonable doubt that Sadio regularly sold the drugs he purchased and purchased more drugs, so he “knowingly and intentionally participated in the narcotics-distribution conspiracy by agreeing to accomplish its illegal objective beyond the mere purchase or sale.”
United States v. Hawkins,
547 F.3d 66, 73-74 (2d Cir.2008).
4.
Conclusion
We have considered all of appellants’ remaining arguments and conclude that they are without merit. For the foregoing reasons, the judgments of conviction of the District Court are AFFIRMED.