PER CURIAM

[DO NOT PUBLISH] Non-Argument Calendar D.C. Docket No. 3:12-cr-00105-MMH-MCR-1 Appeal from the United States District Court for the Middle District of Florida Before JULIE CARNES, JILL PRYOR and FAY, Circuit Judges. PER CURIAM:

John Denton Rouse, Jr. appeals his conviction for possession of at least 500 grams of cocaine with intent to distribute, in violation of 21 U.S.C. § 841(a)(1) and (b)(1)(B). The district court found Rouse guilty at a bench trial. On appeal, Rouse contends that the evidence presented at trial was insufficient, as a matter of law, to prove beyond a reasonable doubt that he possessed at least 500 grams of cocaine because the government mixed the substances recovered from the scene prior to sending them for forensic analysis. After careful review, we affirm.

I. BACKGROUND

A. Charges Against Rouse

A grand jury indicted Rouse on three counts: (1) possession of at least 500 grams of cocaine with intent to distribute (Count 1); (2) possession of at least 28 grams of cocaine base with intent to distribute (Count 2); and (3) possession of a firearm by a convicted felon (Count 3). The district court dismissed Count 2 following a joint motion by both parties. Rouse proceeded to a bench trial on Counts 1 and 3. Rouse was convicted of both offenses, but only challenges his conviction of the Count 1 offense on appeal.

As to Count 1, Rouse stipulated that he knowingly and intentionally possessed cocaine with the intent to distribute it. He disputed only the quantity of cocaine seized and tested by the Drug Enforcement Administration Laboratory ("DEA Lab"), arguing that law enforcement agents improperly commingled bags of white powder into one container before chemically testing whether the bags contained cocaine or pure filler. He argued that this mistake increased the overall weight of the seized narcotics, pushing the amount over the 500 gram threshold for the charged offense.

Rouse did not dispute the weight of cocaine received and tested by the DEA Lab: 195.8 grams of suspected cocaine base (revealed to be powder cocaine) and 859.5 additional grams of powder cocaine. He did not dispute that the 195.8 gram sample was all cocaine. He disputed only how much of the 859.5 gram sample actually was cocaine rather than pure filler. Put differently, at trial Rouse challenged only whether the government could prove beyond a reasonable doubt that at least 304.2 grams of the 859.5 gram sample was cocaine so that the 500 gram threshold for the charged offense was met.

B. Evidence Presented to the District Court

Leading up to the execution of a search warrant at Rouse's residence, detectives from the Jacksonville Sheriff's Office ("JSO") and the DEA directed a controlled purchase of 84 grams of cocaine from one of Rouse's associates. Further investigation revealed that the associate obtained the cocaine from Rouse. JSO detectives executed a search warrant at Rouse's residence and discovered cocaine in a converted garage at the back of the property. Among the detectives was Robert Cook, who had 14 years of experience as a narcotics detective.

Cook testified at trial that, during his years of narcotics experience, he became familiar with certain characteristics of brick cocaine, powder cocaine, crack cocaine, and filler or cutting agents. He described a brick of cocaine as weighing approximately one kilogram, oblong or rectangular in shape, and white or off-white in color. He stated that powder cocaine often appears to have been removed from a block or brick of cocaine, in that it looks "chunked up like it [had been] compressed at some point." Trial Tr., Doc. 171 at 51. Comparatively, Cook testified that crack cocaine is a "darker-color brown" and "a lot of times [its] consistency is going to be a lot harder." Id. at 50. He described filler or cutting agent as a very fine white powder and stated that, in his experience, he had never seen it in brick-like form. He testified that he usually sees filler in jars or plastic bags.

At the scene, Cook found plastic bags of what appeared to him to be either powder or crack cocaine, ranging from what he thought to be one eighth of an ounce up to one ounce in each bag. Cook also found a box of baking soda, which he did not submit to the lab for testing or combine with any suspected cocaine. Photographs of the bags were admitted into evidence by the government and described by Cook at trial.

Some of the bags contained powder of "a little bit rougher consistency" with chunks in it, indicating to Cook that it had been compressed at some point; its appearance was therefore consistent with that of powder cocaine. Id. at 48. Several other bags appeared to Cook to contain crack cocaine based on the substance's color and consistency. Cook separated the bags that appeared to be crack cocaine from those that appeared to be powder cocaine. He did not knowingly combine any filler with the suspected illicit substances. Cook then combined all of the bags of suspected crack cocaine into one container (195.8 grams) and all of the bags of suspected powder cocaine into another container (859.5 grams). He acknowledged that once cocaine was combined with filler he would be unable to discern any difference between cocaine and filler particles. Also, among the bags combined and sent to the lab for testing was a bag labeled "cut-and-dried incense." Id. at 55.

Besides the suspected narcotics, detectives found other drug paraphernalia and contraband at the scene, including multiple gram scales, one of which had visible white powder around the edges. They found approximately $26,600 in U.S. currency and $9,000 in counterfeit currency. Cook testified that a kilogram of cocaine was selling between $25,000 and $30,000 when the cocaine was recovered. Detectives recovered approximately a dozen cell phones, and Cook testified that, in his experience, individuals dealing in larger amounts of cocaine often used several cell phones in an effort to thwart law enforcement's ability to track them through electronic surveillance.

Detectives transferred the two containers of suspected cocaine to the DEA Lab. Carolyn Hudson, a forensic chemist for the DEA, tested samples of each container for cocaine purity. Testing revealed that the 195.8 gram container was 28.6 percent pure powder cocaine and the 859.5 gram container was 5.9 percent pure powder cocaine. Hudson testified that although highly concentrated bricks of cocaine can be as high as 70 to 80 percent pure, the 5.9 percent result was not surprising, and that even with the 5.9 percent finding, she concluded that the entire bag contained cocaine. Hudson also testified that it did not surprise her that Cook originally suspected the 195.8 grams to be crack cocaine because it was moist and chunky.

Hudson identified multiple fillers, also known as adulterants, in the 859.5 gram container, including caffeine, hydroxyzine, lidocaine, and nicotinamide, all of which are loose white or off-white powders. She testified that when these fillers are mixed with cocaine, the entire substance would appear to be about the same. Hudson did not test for baking soda because it is a diluent—which has no effect on the body—and not an adulterant—which does.

C. Defense's Theory and District Court Findings

On cross-examination of Hudson, and during defense counsel's closing argument, Rouse argued that only 84 grams of the 859.5 gram sample of powder cocaine was present at the scene, and that detectives combined the 84 grams with various fillers before the 859.5 gram sample was sent to the DEA Lab for testing. Rouse previously distributed an 84 gram quantity of powder cocaine (about 3 ounces). Defense counsel posed a hypothetical, assuming the 84 grams of powder cocaine was 70 to 80 percent pure. Under that assumption, when the 84 grams were combined with filler to amount to 859.5 grams, the resulting substance was approximately 6.0 percent pure. Rouse argued that this calculation was consistent with the evidence the government presented because the DEA Lab testing revealed that the 859.5 gram bag was 5.9 percent pure.

The defense declined to put on any evidence, and Rouse moved for judgment of acquittal, arguing the government presented insufficient evidence to establish beyond a reasonable doubt that he possessed over 500 grams of cocaine. The district court denied the motion.

The district court found no reason to doubt Cook's testimony that the substances combined to create the 859.5 gram sample were consistent in color and consistency with processed cocaine, and not simple filler. Along with Cook's testimony, the court relied on its review of the photographic evidence showing that the overwhelming majority of the substance tested by the DEA Lab was the chunky, rough consistency of cocaine. The court concluded that this evidence, together with the DEA lab analysis of the actual substance and Hudson's testimony, was sufficient to prove beyond a reasonable doubt that at least 304.2 grams of the 859.5 gram sample was cocaine, thus exceeding—along with the 195.8 gram sample— the 500 gram threshold for the charged offense.

The district court sentenced Rouse to 100 months' imprisonment. This is his appeal.

II. STANDARDS OF REVIEW

We review de novo the denial of a motion for judgment of acquittal based on the sufficiency of the evidence. United States v. Pirela Pirela, 809 F.3d 1195, 1198 (11th Cir. 2015). Under this standard, "[w]e will not reverse a conviction for insufficient evidence in a non-jury trial, unless, upon reviewing the evidence in the light most favorable to the government, no reasonable trier of fact could find guilt beyond a reasonable doubt." Id. at 1198-99 (internal quotation marks omitted). "[A] district court's bench trial findings of fact are reviewed for clear error." Id. at 1199.

III. DISCUSSION

Rouse argues that his conviction should be overturned because the government failed to introduce evidence proving beyond a reasonable doubt that the individual packages of white powder collected at his residence were cocaine before law enforcement commingled the packages into one container. In support, Rouse argues that this Court should adopt a bright-line standard adopted by the Florida Supreme Court in Greenwade v. State, requiring law enforcement to chemically test any substance before combining the contents to meet a minimum statutory threshold if there is an identifiable danger of misidentification. Greenwade v. State, 124 So. 3d 215, 230-31 (Fla. 2013). Rouse also suggests that Cook's testimony was not credible because Cook had misidentified controlled substances here and in the past. We conclude there was sufficient evidence to support Rouse's conviction.

First, we cannot adopt the bright-line rule from Greenwade because it conflicts with our prior precedent in United States v. Baggett, 954 F.2d 674, 677-78 (11th Cir. 1992). In Baggett, we reinforced this circuit's "expansive view" that controlled substances can be identified by various means of circumstantial evidence, such as "lay experience based on familiarity through prior use, trading, or law enforcement; a high sales price; on-the-scene remarks by a conspirator identifying the substance as a drug; and behavior characteristic of sales and use, such as testing, weighing, cutting and peculiar ingestion." Id. at 677 (internal quotation marks omitted).

An exception under Greenwade would require this court to adopt a standard requiring chemical testing of certain controlled substances, those that present an "identifiable danger of misidentification," prior to combining suspected narcotics. Greenwade, 124 So. 3d at 231. Because Baggett allows identification through circumstantial evidence for all controlled substances, application of the Greenwade exception would swallow the Baggett rule. Baggett, 954 F.2d at 677. Even though the rule in Greenwade may encourage best police practice, our prior precedent leaves no room for the exception Rouse urges, and we must therefore consider the findings of the district court in light of this precedent. Id.

The district court determined that Cook's uncontested testimony, which was based on his 14 years of narcotics experience, and the photographic evidence showing that the overwhelming majority of the substance tested by the DEA Lab was the chunky, rough consistency of cocaine was sufficient to conclude, beyond a reasonable doubt, that Rouse possessed at least 500 grams of powder cocaine. Cook's testimony and the photographic evidence fall within the acceptable types of circumstantial evidence that courts may rely upon to determine whether a substance is cocaine. Id. In reviewing the district court's finding from this acceptable form of circumstantial evidence, viewed in the light most favorable to the government, we find no clear error. Pirela Pirela, 809 F.3d at 1199.

Second, regarding Cook's testimony, the credibility determination was the exclusive province of the district court, and we do not find the testimony incredible as a matter of law. See United States v. Thompson, 422 F.3d 1285, 1291-92 (11th Cir. 2005). To be incredible as a matter of law, testimony must be unbelievable on its face, i.e., testimony as to facts that the witness could not have possibly observed or events that could not have occurred under the laws of nature. Id. at 1291 (alteration adopted) (internal quotation marks omitted). It is undisputed that Cook was present at the execution of the search warrant, he observed the cocaine samples in person, and his assessment of the samples based on his 14 years of narcotics experience was not incredible as a matter of law.

The evidence, when viewed in the light most favorable to the government, was sufficient to establish that Rouse knowingly and intentionally possessed at least 500 grams of cocaine with the intent to distribute.

IV. CONCLUSION

For the reasons set forth above, we affirm the judgment of the district court.

AFFIRMED.