Opinion by Judge Schroeder.
Dissent by Judge Reinhardt.
OPINION
SCHROEDER, Circuit Judge:
Government and commercial entities enjoy increasing capacity to obtain, store, and analyze information about people, giving rise to increasing concerns about privacy. Nowhere is that upward spiral more evident than in litigation, like this case, calling into question practices relating to identifying people through their deoxyribo-nucleic acid (DNA) sequences.
The appellant here, Thomas Kriesel, pled guilty to a drug conspiracy charge, and was sentenced to a term of imprisonment followed by a term of supervised release. One condition of his supervised release required him to provide a blood sample for analysis of his DNA, and inclusion of his DNA profile into the government’s Combined DNA Index System (CO-DIS) database. Now that Kriesel has completed his term of supervised release, he has made an unusual invocation of Federal Rule of Criminal Procedure 41(g), asking the government to return the blood sample. He has no remaining objection to the government’s retention of the information in his DNA profile, which the government analyzed from the extracted blood sample, and which it currently stores in the CODIS database.
Kriesel argued to the district court that the government had no legitimate reason for retaining the blood sample — which of course has within it not only the limited information the government has analyzed for his DNA profile, but his entire unana-lyzed genome. The district court ruled the government had a legitimate purpose in retaining the blood samples that generate the CODIS profiles in order to ensure that the matches to forensic evidence, identified through CODIS searches, are accurate. The court found no reason at this time to believe the government would use the blood for other purposes, many of [1140]*1140which are already prohibited by statute. The district court therefore granted judgment to the government, and we affirm on a similar basis.
I. THE GOVERNMENT’S DNA DATABASE
The Federal Bureau of Investigation (FBI) administers CODIS as a nationwide database of genetic identifying information.
The CODIS database stores DNA profiles of convicted federal felons on supervised release and others who have had brushes with the law. See DNA Analysis Backlog Elimination Act of 2000 (DNA Act), Pub.L. No. 106-546, § 3, 114 Stat. 2746, 2728-30; see also 28 C.F.R. § 28.2. These DNA profiles are commonly generated from blood samples.
The blood is collected from offenders and then sent to the Federal DNA Database Unit (FDDU) in Quantieo, Virginia. The FDDU extracts the DNA molecules from each blood sample, analyzes the molecules, generates a profile of identifying characteristics, and uploads the profile to CODIS. In addition to storing the profiles in CODIS, the FDDU retains offenders’ physical blood samples to help ensure accurate matches to DNA found at crime scenes.
It is important in this case to understand how the government uses both the DNA profile and the samples for identification purposes. Blood cells in the samples contain two types of DNA: the biologically important coding (or non junk) DNA, and the biologically unimportant non-coding (or junk) DNA. See United States v. Kincade, 379 F.3d 813, 818 (9th Cir.2004) (en banc) (plurality) (explaining the difference between junk and non junk DNA). We held in Kincade that the government may extract junk DNA from samples, and use it to generate profiles for inclusion in CODIS, because present scientific understanding indicates that junk DNA reveals no sensitive, private genetic or medical information. It is useful, however, for identification purposes. See id. The government uses only junk DNA to generate the CODIS profile. The record in this case reflects that the government makes no use of the non junk DNA in the blood sample.
The CODIS system searches for matches between offenders’ DNA and crime scene evidence. It is when a match is found that the actual sample is tested. The federal lab retrieves the offender’s actual blood sample, which it has retained in storage. It again extracts junk DNA from that sample, generates a new DNA profile, and compares the new profile to the CODIS profile. This verifies that the person whose profile CODIS matched to the crime scene evidence is the same person who provided the original blood sample.
The FBI created CODIS after Congress passed the Violent Crime Control and Law Enforcement Act of 1994, which authorized the agency to create a national database of DNA samples from convicted federal offenders. See Pub.L. No. 103-322, 108 Stat. 1796 (Sept. 13, 1994). Following the creation of CODIS, all fifty states passed laws requiring convicted felons to provide DNA samples for CODIS.
The legislative history reveals concerns that the database be as accurate and trustworthy as possible. Representative Henry Hyde acknowledged that while a DNA database could be valuable to law enforcement, the accuracy of the database was critical:
[Wjhen properly performed, DNA analysis has proven an extremely effective investigative tool in the criminal justice process. As DNA technology is increasingly employed in the courtroom, however, there has been growing concern over [1141]*1141issues of quality assurance and standards for conducting DNA testing. H.R. 829 [the DNA Act] will guarantee that needed quality assurance standards are developed and implemented.
139 Cong. Rec. H1650-01 (daily ed. Mar. 29,1999) (statement by Rep. Hyde).
In response to such concerns over the accuracy of CODIS, the 1994 Act requires the Director of the FBI to develop quality assurance and proficiency testing standards. 42 U.S.C. § 14131(a)(C)(2). Recommendations for these standards were to come from “an advisory board on DNA quality assurance methods from among nominations proposed by the head of the National Academy of Sciences and professional societies of crime laboratory officials.” Id. at § 14131(a)(1)(A). Responsibility for implementing the standards lies with the FBI. The statute provides that the FBI Director “after taking into consideration such recommended standards, shall issue (and revise from time to time) standards for quality assurance, including standards for testing the proficiency of forensic laboratories, and forensic analysts, in conducting analyses of DNA.” Id. at § 14131(a)(C)(2).
The procedures challenged in this case thus came from the advisory board’s recommendations. The FBI Director implemented “Quality Assurance Standards” in which he required that “where possible” the actual blood samples of offenders be retained in the CODIS database. Fed. Bureau of Investigation, Quality Assurance Standards for DNA Databasing Laboratories (revised July 1, 2009), available at http://www.fbi.gov/about-us/lab/codis/ qas_databaselabs.pdf. The FBI thus retains Kriesel’s blood sample as part of its implementation of quality and accuracy standards developed pursuant to Congressional directives.
II. THIS LITIGATION’S HISTORY
In this round of Kriesel’s extended litigation, his Rule 41(g) motion asks us to order the government to return his original blood sample. The litigation has a long history and has already been to this court twice. See United States v. Kriesel, 508 F.3d 941 (9th Cir.2007) (Kriesel I); United States v. Kriesel, 604 F.3d 1124 (9th Cir.2010) (Kriesel II).
Kriesel pleaded guilty in 1999 to one count of conspiracy to commit possession with intent to distribute methamphetamine. 21 U.S.C. §§ 841(a)(1), 846. He was sentenced to a term of 30 months imprisonment, followed by a term of 36 months supervised release. As a condition of his supervised release, the probation office required him to provide a blood sample for DNA analysis. Kriesel objected on the ground that the requirement was unlawful, and the probation office asked the district court to revoke his supervised release on account of his refusal to provide the sample.
The district court rejected Kriesel’s arguments, which were, first, that the extraction of a blood sample for DNA analysis violated the Fourth Amendment, and second, that the implementing regulations were promulgated in violation of the Administrative Procedure Act (APA), 5 U.S.C. § 553. The court revoked his supervised release.
A. Kriesel I
Kriesel appealed to this court, and we affirmed. On Kriesel’s Fourth Amendment challenge, we considered the totality of the circumstances, and viewed Kriesel’s status as a convicted felon on supervised release to be key for our analysis. Noting Kriesel’s “diminished expectation of privacy” as a supervised releasee, we concluded he had only a limited interest in preventing the government from confirming his [1142]*1142identity through analysis of his junk DNA. See Kriesel I, 508 F.3d at 947 (citing Kincade, 379 F.3d at 833 (plurality)). We recognized that Kriesel raised a number of legitimate concerns about other DNA tests, i.e., “about DNA samples being used beyond identification purposes.” Id. at 948. The presence of significant criminal prohibitions on the unauthorized use or disclosure of DNA samples, see 42 U.S.C. § 14135e(c), however, supported our conclusion that Kriesel’s concerns, on the record before us, were speculative.
Our majority opinion in Kriesel I recognized three “undeniably compelling” interests that supported the collection and analysis of DNA samples of convicted felons: (1) linking supervised releasees to crimes committed while on release; (2) deterring future crimes; and (3) solving past crimes. Id. The majority also rejected Kriesel’s APA challenge, concluding the implementing regulations were exempt from the notice-and-comment process. See id. at 945 (citing § 553(b)(3)(A)). We did not decide what information could be retained after Kriesel served his term of supervised release, or as Judge Gould had asked in his concurrence in Kincade, when supervised releasees had “wholly cleared their debt to society.” Id. at 949 (quoting Kincade, 379 F.3d at 841 (Gould, J., concurring in the judgment)).
After our 2007 decision, Kriesel submitted to DNA extraction by providing the blood sample to the district court’s probation office. His sample was analyzed, the profile uploaded to CODIS, and the sample stored with the FDDU. Kriesel thus complied with all conditions of his sentence.
Kriesel then filed a motion in the district court asking, on Fourth Amendment grounds, that his DNA profile stored in CODIS be removed, and also seeking under Fed.R.Crim.P. 41(g) return of the blood sample. The district court denied the motion, concluding that while a former supervised releasee had more privacy interests than a current supervised releasee, on balance the government’s interests in retaining the CODIS profile outweighed Kriesel’s privacy interests. The court did not specifically address the Rule 41(g) claim that is now before us.
B. Kriesel II
On appeal for the second time, Kriesel abandoned his Fourth Amendment argument for expungement of his DNA profile stored in CODIS. He argued only that he was entitled to the return of the blood sample as his property under Rule 41(g). See Kriesel II, 604 F.3d at 1124-25. We concluded that Kriesel’s Rule 41(g) claim had been so subsumed within his now-abandoned Fourth Amendment argument, that the case should be remanded on an open record for the district court to specifically address and determine whether Kriesel was entitled to the return of his blood sample under Rule 41(g). Id.
C. Proceedings in the District Court Underlying This Appeal
On remand from our decision in Kriesel II, the parties developed the evidentiary record pertaining to the blood samples. Declarations explained how DNA is extracted from blood samples, and how the samples are used. The district court considered this evidence, and after a hearing, it denied Kriesel’s Rule 41(g) motion.
The record developed on remand described in detail the many steps involved in the process of collecting blood samples from known offenders, and generating profiles from the junk DNA for inclusion in CODIS. We describe only the general process the FDDU follows, for specifics may differ from lab to lab in state jurisdictions.
The first step is a government agent’s collection of a blood sample from a known [1143]*1143offender. The agent places several drops of blood on each of two halves of a special paper card used for storage. The agent marks the card with information identifying the offender who provided the sample. The agent submits the card to the FDDU with Form FD-936, “Request for National DNA Database Entry.” That form contains the offender’s fingerprints as well as information about the offender and the submitting agency. The paper card and the form are thus sent together to the FDDU for processing.
In the second step, the FDDU prepares the sample for analysis and eventual inclusion of the profile in CODIS. The FDDU ensures the samples have not been contaminated during transit, then assigns the sample a unique “specimen identification number.” This number allows the government to track the sample throughout the analysis process.
The FDDU processes the sample, after separating the halves of the paper card used for transport. One half is treated with chemicals that aid DNA analysis by breaking down blood cells and stabilizing the DNA molecules in a way that permits long term storage at room temperature. The other half of the card is not treated with chemicals, and contains unadulterated DNA molecules; this half must be kept and stored separately in a freezer. The lab keeps both sides of the card for DNA identification purposes.
The FDDU uses the treated side of the card to generate the DNA profile. The lab identifies molecular markers throughout the extracted junk DNA, and turns this information into a searchable database entry that is uploaded to the CODIS database. The database entry itself contains specimen identification information linking the profile to the stored blood sample, but not the subject’s name or any other personally identifiable information. Personal identifying information such as the offender’s name, social security number, and fingerprints are contained in the Form FD-936, which the FBI stores in a secure filing facility.
The primary value of CODIS is to link unidentified DNA samples collected from crime scenes to the DNA of known offenders in the system. See Kriesel I, 508 F.3d at 949. Investigators collect forensic samples from crime scenes and unidentified human remains, or from the relatives of missing persons who volunteer samples to aid the search. See, e.g., 42 U.S.C. § 14132(a)(2)-(4) (authorizing generation of DNA profiles from these sources).
Forensic laboratories around the country analyze DNA from those samples to generate a profile that can be matched to known offenders’ DNA. The process produces accurate identifications because it is extremely unlikely two people will have the same profile. See Kincade, 379 F.3d at 818-19. The goal of the CODIS system is to find what the government calls a “Candidate Match” — a putative match between an identified offender and unidentified crime scene DNA. When CODIS finds a Candidate Match, the FDDU receives a CODIS Match Report.
At this point the original blood sample is used for confirmation. Upon receiving a CODIS Match Report, the FDDU retrieves the original blood sample from storage to perform further analysis — what the federal government has called in this case the “Match Confirmation” step. The FDDU locates the retained blood sample in storage using the specimen identification number. The FDDU then takes the retained sample, re-extracts junk DNA from it, and runs a new analysis. If the newly generated profile is the same as the one in CODIS that formed a Candidate Match, the match is confirmed and the accuracy of the match between the CODIS profile and the identified offender is en[1144]*1144sured. The confirmation of the CODIS match is thus achieved by comparing the profile generated from the retained sample with the Codis profile. Although CODIS has not yet encountered such a “mismatch” or “misidentification” error, in the event that the generated profile did not match the CODIS profile, the lab would then determine what caused the error and, presumably, prevent similar errors from occurring in the future. The “Match Confirmation” is used to ensure the continued accuracy and integrity of the CODIS system. It is the government’s primary justification for retaining the blood sample.
In deciding the Rule 41g issues, the district court first ruled that the blood sample was property within the meaning of the Rule, and that Kriesel, because of his concerns about the private information the sample’s non junk DNA in the sample could reveal, was sufficiently aggrieved to seek its return. The district court denied the Rule 41(g) motion, however, because the government had shown a sufficient reason for retaining the sample. It ruled that the use of the “Match Confirmation” process to ensure the accuracy of the system satisfied the government’s burden to justify retention. The court observed that Kriesel’s sample was an integral part of the database, and that Rule 41 should not be permitted to undermine it. The court recognized that if Kriesel and other offenders whose DNA profiles were included in CODIS were able to petition successfully for the return of their blood samples, then “offenders’ identities would be released without Match Report confirmations” and “CODIS’s integrity would erode.” United States v. Kriesel, No. CR03-5258, at *11-12 (W.D.Wash. July 21, 2011).
The government offered a number of other justifications for retaining the blood sample, including a speculative interest in being able to utilize as yet undeveloped technology. The district court properly rejected these justifications, and they are not seriously at issue in this appeal.
Kriesel timely appeals. His primary argument is that the district court erred in concluding that retaining his blood sample is reasonable. Although the district court did so because retention allows the government to ensure the accuracy of putative matches between offenders and unidentified DNA, Kriesel contends the “Match Confirmation” process is not a reasonable justification because it is not necessary to ensure the accuracy of Candidate Matches. He urges us to rule that those matches are sufficiently reliable without resort to the confirmation process, so as to eliminate any reason for retaining the sample. He also argues the federal government does not need to retain the sample if some states, such as Wisconsin, are not required to retain samples as a condition of participating in CODIS.
III. DISCUSSION
Rule 41(g) provides the textual basis for Kriesel’s claim for return of his blood sample. The rule provides “[a] person aggrieved ... by the deprivation of property may move for the property’s return.” We have held that a defendant’s Rule 41(g) motion should presumptively be granted if the government “no longer needs the property for evidence.” United States v. Fitzen, 80 F.3d 387, 388 (9th Cir.1996) (citation omitted). The government can rebut that presumption, however, by showing a continued need for the property that is reasonable under all of the circumstances. Ramsden v. United States, 2 F.3d 322, 326 (9th Cir.1993) (citing Fed.R.Crim.P. 41, Adv. Comm. Notes to 1989 Amendments).
The district court correctly held that Kriesel is seeking the return of “property.” The Rule’s definition of property [1145]*1145includes “documents, books, papers, any other tangible objects, and information.” Fed.R.Crim.P. 41(a)(2)(A). The district court also properly concluded that the blood sample itself is a tangible object, and the genetic code contained within the blood sample is information. The applicability of Rule 41 to bodily fluids is supported by our circuit law. We have previously held that professional baseball players’ urine samples, that the government seized from a laboratory, were “property” within the meaning of Rule 41(g). United States v. Comprehensive Drug Testing, 621, F.3d 1162, 1173 (9th Cir.2010).
The district court in this case further concluded Kriesel was “aggrieved,” within the meaning of the Rule, because the government has retained the sample from which private information can be extracted. We agree this is sufficient to allow him to bring a Rule 41(g) motion. We therefore reject the government’s contention, raised for the first time on appeal, that he lacks standing.
The issue under Rule 41(g) is thus whether the government has shown a legitimate reason for retaining the property. This case turns on whether the government’s continued retention of the blood sample is “reasonable[ ] under all of the circumstances.” Ramsden, 2 F.3d at 326 (citation omitted). The district court found that the use of the blood samples to ensure the accuracy of DNA identification was a valid reason to retain the sample. The issue is one we review de novo. See United States v. Kaczynski, 416 F.3d 971, 974 (9th Cir.2005).
The government explained to the district court how and why it uses the blood sample to ensure it has accurately determined the identity of the person associated with the DNA profile stored in CODIS. This “Match Confirmation” process compares the database profile against a new profile generated from the offender’s retained blood sample. This ensures that the system is working accurately. It also enables pre-arrest confirmation of a match. The dissent is incorrect in suggesting the same purpose is served by testing the subject after he has been hauled into custody.
The Match Confirmation process is also a method of long-term quality control. The government’s quality assurance standards require periodic random testing every six months of previously analyzed samples. The FDDU therefore combines random samples processed in the intervening six months, together with any positive Match Confirmation samples, and generates new DNA profiles from these samples. This is done for at least 1% of the total number of samples processed in the intervening six month period. By comparing the new and old profiles, the lab can identify errors in the DNA analysis process. There remains a risk, however, that some of the untested remainder of samples (up to 99%) were analyzed erroneously. Retaining blood samples for the Match Confirmation process increases the chances the government will catch any such errors. If the Match Confirmation were to reveal an error in the DNA analysis process, the lab could review how the process might have malfunctioned. The statute reflects the Congressional concerns, as illustrated by Representative Hyde’s remarks, in maintaining the accuracy of the system. The statute expressly requires the FBI to implement standards to maintain laboratory accuracy. 42 U.S.C. § 14131(a)(C)(2). The “Match Confirmation” process carries out that Congressional mandate.
Kriesel does not quibble with the government’s argument that this confirmation process avoids generating false identifications. He argues, rather, that as a procedural matter the government does not need such a rigorous means of quality [1146]*1146assurance, since the record indicates the government has never found an erroneous match between a DNA profile and a blood sample. Kriesel thus contends that if the system has been flawless in the past, there is no need to retain the blood samples.
A strong record of quality assurance in the past, however, is not necessarily a reason to abandon protections that will ensure accuracy in the future. The argument is a bit like contending a driver should stop carrying a spare tire because the car has never had a flat. Moreover, as the district court correctly noted, the retention of the blood samples promotes the efficacy of the CODIS identification system and maintains the credibility of the system, should CODIS ever provide a false match. The double-check helps maintain the confidence of the public and law enforcement in CODIS as a means of recreating genetic profiles in the event that the CODIS profiles ever became compromised. We therefore agree with the district court that the government carried its burden of showing the samples were retained for a reasonable use.
Kriesel points out that some laboratories participate in the nationwide network of DNA databases without retaining the samples. He cites Wisconsin as an example, where state law requires the lab to “destroy specimens obtained ... after analysis has been completed.” Wis. Stat. § 165.77 (2012). Kriesel contends that because the federal government nevertheless permits Wisconsin to participate in CO-DIS, the FBI’s retention of the sample must be unnecessary.
Federal sufferance of the Wisconsin law, however, does not serve to invalidate the federal retention policy. Federal statutes neither require nor forbid retention by the states of blood samples as part of the administration of CODIS. Federal law does require the federal government to implement quality control standards and thus authorizes the federal government to adopt more stringent measures than states participating in CODIS. 42 U.S.C. § 14131(a)(C)(2).
Because the government obtained Kries-el’s sample as a condition of his supervised released from custody, the remaining question under Rule 41 is whether the government’s justification for retention of the sample disappeared when Kriesel completed the term of supervised release. The question as originally framed by Judge Gould in his concurrence in Kincade was whether the record of a felon’s DNA could be retained once those formerly on supervised release had “wholly cleared their debt to society.” Kincade at 813 (Gould, J., concurring).
This query did not distinguish between the CODIS profile and the blood sample, but asked whether any DNA record need be retained after completion of a sentence. In this case, Kriesel’s completion of his sentence did not eliminate the government’s interest in retaining the DNA profile in CODIS, and Kriesel no longer argues that it did. The need to identify DNA found at a crime scene is critical and accuracy matters. See Kriesel I, 508 F.3d at 949-50. The retention of the blood samples further those goals by ensuring the accuracy of the CODIS profile match, which is the function served by the “Match Conformation.”
Kriesel has abandoned his Fourth Amendment argument, so we do not separately address whether completion of the sentence implicates additional Fourth Amendment privacy concerns. This case does not raise the concerns that have been expressed about the government’s taking of DNA from persons who have not been convicted of any crime. See Maryland v. King, — U.S. -, 133 S.Ct. 1958, 1989, 186 L.Ed.2d 1 (2013) (Scalia, J., dissenting); Haskell v. Harris, 686 F.3d 1121 (9th Cir.2012) (vacating panel opinion and or[1147]*1147dering rehearing en banc); Haskell, No. 10-15152, Dkt. No. 111 (Nov. 13, 2012) (deferring submission to en banc court pending the Supreme Court’s decision in King).
The dissent focuses on the sensitive personal information contained within DNA that can be obtained from a blood sample, echoing concerns of the dissenters in Kin-cade. There is no basis in this record, however, to conclude that the government will even try to obtain such information. The DNA Act, with certain limited exceptions not relevant here, proscribes the government from using a DNA sample for any purpose other than suspect identification. See 42 U.S.C. § 14132(b)(3). Any person who knowingly uses a DNA sample or result without authorization is subject to criminal penalties. Id. at § 14135e(c). There is nothing to suggest that the government will use Kriesel’s blood sample for anything more than identification purposes, which of course it is permitted to do. We therefore decline to credit the speculative concerns he raises.
In dealing with speculative concerns about DNA, we have previously stressed that we must look only to the record on appeal and what it shows about the actual scope and uses of the DNA information. See Kincade, 379 F.3d at 838. Other courts have similarly refused to speculate about the uses to which the government could conceivably put DNA information. See, e.g., United States v. Weikert, 504 F.3d 1, 14 (1st Cir.2007); United States v. Amerson, 483 F.3d 73, 87 (2d Cir.2007); United States v. Mitchell, 652 F.3d 387, 408 (3d Cir.2011). Indeed, only recently the Supreme Court has held that plaintiffs lacked standing to challenge the constitutionality of the FISA Amendments Act of 2008, 50 U.S.C. § 1881a. Clapper v. Amnesty Int’l USA — U.S. -, 133 S.Ct. 1138, 1150, 185 L.Ed.2d 264 (2013). The Court stated in Clapper that the plaintiffs’ claims that they would be the target of illegal surveillance were speculative, and thus did not constitute an injury-in-fact. Id. at 1148-49. The court reasoned that the “injury” the plaintiffs complained of was predicated on speculation about what actions the government might take in the future and lacked support in the record. Id. A fortiori, if the plaintiffs in Clapper lacked standing to complain of injury that was only speculative, then Kriesel’s demand for return of property lacks merit when founded upon similar speculative concerns about possible future government conduct.
We nevertheless recognize that we are dealing with a rapidly changing world in which risks of undue intrusions on privacy are also changing. We have previously stressed that if scientific discoveries make clear that junk DNA reveals more about individuals than we have previously understood, we should reconsider the government’s DNA collection programs. See, e.g., Kriesel I, 508 F.3d at 947 (“Should the uses to which junk DNA can be put be shown in the future to be significantly greater than the record before us today suggests, a reconsideration of the reasonableness balance struck would be necessary, even with respect to individuals in Kriesel’s position.” (citation and internal quotation marks omitted)).
We affirm the district court’s holding that the government’s continued retention of Kriesel’s blood sample is reasonable under the circumstances presented on this record.
The district court’s denial of Kriesel’s Rule 41(g) motion is AFFIRMED.