Case Information
UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA UNITED STATES OF AMERICA, Case No. 17-cr-00533-EMC-1 Plaintiff, ORDER DENYING DEFENDANTS’ v. MOTION TO EXCLUDE OR LIMIT
“CAST” TESTIMONY JONATHAN JOSEPH NELSON, et al., Docket No. 1444 Defendant. I. INTRODUCTION
Pending before the Court is Defendant Brian Wendt’s motion to exclude or limit the
proposed testimony of Special Agent Meredith Sparano. See Docket No. 1444. Ms. Sparano is a member of the FBI’s Cellular Analysis Survey Team (“CAST”) whose testimony the Government proffers as expert opinion on cell phone communications and “[h]istorical cell site data analysis.” Docket No. 1490-1 (“Sparano Decl.”) ¶ 4. In essence, Ms. Sparano will testify as to how cell phone call detail records (“CDRs”) reveal the general geographic locations of cell phones at particular points in time based on the location of the cell towers with which the phones are communicating. Ms. Sparano will also illustrate tower locations relevant to the instant case on illustrative maps in a PowerPoint presentation.
Defendants do not take issue with the accuracy and use of CDRs, nor do they dispute the
way that cell phones connect with particular cell towers. Instead, they focus on the admissibility
of Ms. Sparano’s mapping testimony. Mr. Wendt argues that Ms. Sparano’s testimony should be
excluded or limited for several reasons, including (1) that the Government’s disclosures about its
proprietary mapping software do not satisfy Federal Rule of Criminal Procedure 16 or this Court’s
discovery order of January 13, 2021; (2) that Ms. Sparano is not qualified to offer expert opinion
in reliance on the Government’s proprietary mapping software for purposes of Federal Rule of
Evidence 702 and
Daubert v. Merrell Dow Pharmaceuticals, Inc.
,
A. Procedural History
As the Government points out, the parties have previously engaged in “two rounds of litigation on CAST.” Docket No. 1489 (“Opp’n Br.”) at 1. On November 29, 2020, Defendants filed a motion (in the form of a joint defense letter request) with Magistrate Judge Beeler for discovery regarding Ms. Sparano’s proposed testimony pursuant to Federal Rule of Criminal Procedure 16(a)(1)(G). Docket No. 1337. The Government opposed Defendants’ motion on December 7, 2020, noting that it had earlier provided Defendants with the declaration by Ms. Sparano that continues to serve as the foundation of her testimony on August 14, 2020. See Docket No. 1356 at 1 n.1; Docket No. 1490-1. In an order of December 20, 2020, Judge Beeler denied Defendants’ requests for additional discovery concerning the factual bases of Ms. Sparano’s opinions. See Docket No. 1378. Judge Beeler concluded that there was “no more information to disclose” about Ms. Sparano’s methodology beyond what she recited in her declaration and that Defendants were free to “challenge the sufficiency of the predicates for [her] expert opinion through the[ir] contemplated Daubert motion.” [1] at 2.
On December 31, 2020, Defendants filed an appeal of Judge Beeler’s discovery order with this Court. See Docket No. 1388. The Court addressed the issue at the parties’ status conference of January 13, 2020. In response to Defendants’ complaint that they had not been provided adequate information about the “particular software program” used by Ms. Sparano to prepare her PowerPoint presentation, the Court ordered the Government, out of an abundance of caution, to identify the program and “to provide sufficient information ( e.g. , in the form of user manuals, technical guides, etc., or descriptions by the [Sparano] of specific steps taken) to determine how the program was used.” Docket No. 1419 at 2. The Court further instructed that if the Government wished to prove the reliability of Ms. Sparano’s methodology by reference to any peer review of the PowerPoint presentation it had to disclose to Defendants “the protocols underlying the peer review process (including software used by the reviewers).” Id. The Government sent Defendants a discovery letter on January 15, 2021, identifying its software program, called ESPA, and describing its peer review process. [2] Docket No. 1547-1. In response to Mr. Wendt’s motion, and Mr. Ott’s and Mr. Diaz’s supplemental brief in support thereof, the Government also filed Ms. Sparano’s declaration (which had previously been lodged with the Court the previous month; see Docket No. 1396-4) as well as the then-current version of Ms. Sparano’s PowerPoint slideshow, dated December 18, 2020. [3] See Sparano Decl., Docket 1490-2 (“Slideshow”).
The Court held a three-hour hearing on Defendants’ motion on March 2, 2021. Docket
No. 1562 (“Hearing Tr.”). The Court took the matter under submission.
B. Proposed Testimony
As noted above, the foundation of Ms. Sparano’s proffered testimony is contained in her declaration of August 12, 2020. The declaration summarizes Ms. Sparano’s training, experience, and methodology as a member of CAST. [4] In it, Ms. Sparano states that she has “served as a Special Agent for the FBI since 2015” and has been assigned to CAST for “the past three years.” Sparano Decl. ¶ 1. Regarding her qualifications to testify as an expert, Ms. Sparano claims to “have over 500 hours of training in the area of historical call detail records analysis” and “cellular phone and cellular network technology.” Id. Her background “includes training from the FBI, other U.S. Government Agencies, engineers from the Florida Institute of Technology, and engineers from all the major cellular network providers.” Id. Ms. Sparano herself also “conduct[s] training regarding cellular technology and record analysis for various state, local, and federal law enforcement personnel.” ¶ 2. Further, she has “been qualified as an expert on the analysis of historical cell site data on one occasion” in California state court and she represents that her analysis of such data “has been utilized on multiple occasions in federal and state courts throughout the United States.” [5] ¶ 3.
Ms. Sparano then discusses the scientific principles underlying cell site location information and the methodology customarily employed by CAST agents.
4. Historical cell site data analysis is based on the premise that the approximate (not exact) location of a cell phone can be determined by examining the records generated by the cell phone when it was used. These records are called CDRs (call detail 1 records). When a cell phone communicates with a cell tower (voice, text message[,] or data session), a record is produced showing the 2 cell tower and sector the phone utilized to complete the call, text[,] or data session. The cell tower has a specific number combination 3 that is unique to that geographic area (no other tower within the network will have these same numbers). All of the service providers 4 maintain a “tower list”, archived periodically throughout the year, which provide the exact location of towers within their network. 5 The tower list contains additional information that is used to determine the orientation of the tower sectors. . . . 6 5. Three things are needed to conduct historical record
7 analysis[:] a set of CDRs, a tower list, and address information/ facts concerning the crime. The CAST team illustrates this information 8 onto a mapping platform similar to Microsoft Map Point or Google Earth to illustrate the approximate location of the phone. The 9 illustration of the tower and sector is commonly represented by a 10 symbol known as an open wedge. The wedge shows the direction radio frequency is moving away from the tower along the center line of the [sector]. [6] These shapes have been used since the beginning of historical cell site analysis by cell phone service providers, who advocate the use “wedges” to illustrate orientation and can demonstrate the potential coverage area of a cell site. In some cases, service providers may have additional data they can provide law enforcement . . . to further define the approximate location of a cell phone. 6. After plotting relevant events from the CDRs on a map, address information and known facts are then considered. Results and/or conclusions are compiled in an official report and/or presentation for use by investigators or at trial. The maps in the official report and/or presentation demonstrate the general area where the cellular phone associated with the CDR was located on the date and at the time the records were generated. ¶¶ 4-6 (emphasis added).
Ms. Sparano next explains that CAST agents can verify the actual coverage area of a cell tower ( i.e. , beyond the potential coverage area indicated by the CDRs) in numerous ways. These include “cell tower surveys” and “drive tests.” [7] With cell tower surveys, CAST agents employ an engineering phone and “go to certain points within the projected area of coverage (within the [open] wedge) and take measurements to corroborate the coverage area of a particular cell sector.” Id. ¶ 7. With drive tests, agents begin by “identifying the towers of interest (target towers) and a target area,” which is “defined on a map” and “encompasses the target towers as well as ‘peripheral towers.’” Id. ¶ 8. Agents then take surveying equipment to the target area and drive the equipment “in a grid search pattern” so as “to measure and record data from the target and peripheral towers.” Id. These data are finally “processed and displayed on maps to demonstrate the actual coverage area of the target tower.” Id. Ms. Sparano states that both cell tower surveys and drive tests “have been utilized over the past ten years by CAST Agents to verify the accuracy of theoretical illustrations”— i.e. , those generated solely from CDR information. See id. ¶ 9. She claims that while “these methods typically refine the footprint created by a cell tower” they generally “support the use of historical cell site data to illustrate the approximate location of a phone at the time a call is generated.” Id. ; see also id. ¶ 11 (“Based on my experience and the collective experience of the FBI CAST Team, historical record analysis has repeatedly proved highly effective at determining the approximate location of cell phones at the time calls, text messages, or data sessions are conducted and the CDRs are generated.”). Ms. Sparano then discusses the mechanics by which cell phones connect to cell towers based on her “training from cellular network engineers.” ¶ 10. She explains that “cell phones are designed to select which cell tower to communicate with for call origination . . . based on one overriding factor,” namely which tower has “the strongest, best quality signal” at a particular location and particular time. (emphasis added). The tower with the clearest signal, she clarifies, might not be the tower that “was the closest to the phone at the time the call was generated.” Id. CDRs thus indicate “the cell tower that was, in fact, chosen by the phone” because it “provided the best service” to the phone at a given moment, “regardless of whether that particular cell tower was the closest to the phone at the time the call was generated.” Id. ; see also id. (“Although the cell tower listed in the CDRs is not always the closest tower, the phone must physical evidence, GPS devices, etc.” Sparano Decl. ¶ 12.
have been within the ‘footprint’ of the cell tower that appears in the call detail record.”). As a result, a CDR “will provide the approximate location—not the exact location—of the cell phone at the time the record was generated.” (emphasis in original).
Ms. Sparano’s declaration concludes by describing her specific methodology in the instant case.
13. In this case, I conducted historical cellular analysis in accordance with the methods laid out above. First, I was provided sets of CDRs and address information/facts concerning the crime, specifically a suspected homicide that occurred on or about July 15, 2014, in the vicinity of Fresno, California. I understand that the CDRs provided to me were obtained from search warrants executed in this case for a set of nine phone numbers of interest in the suspected homicide. The CDRs contained information regarding the towers accessed for each phone call, text message, and data session. I also obtained the tower list from the cell phone provider for the relevant time period. I then analyzed the records and compiled a report, which depicted the cellular activity of the phone numbers of interest relative to significant times and locations of the suspected homicide. As part of creating the report, I imported the CDRs and tower information into a mapping program to obtain a visual depiction of the tower locations accessed by each target phone number during relevant times to the crime . No drive test was performed in this case. ¶ 13 (emphasis added). As noted above, Ms. Sparano’s proffered testimony will also be conveyed through the PowerPoint presentation that the Government filed contemporaneously with her declaration. See Slides. The presentation, in its current iteration, includes sixty slides, the vast majority of which feature spreadsheets with underlying CDR data and/or mapping images generated through the Government’s ESPA software. See id. The mapping images typically depict a particular location ( e.g. , the area around Fresno, California) on which are superimposed open wedges (indicating cell tower locations and sector orientations) and tables listing the time(s) at which a target phone connected to the relevant towers. See id. Some of the images also contain additional graphics such as arrows indicating a purported direction of travel, see, e.g. , id. at 20, or explanatory inferences concerning the CDR data captured elsewhere on the slides, id. at 39. The slides do not identify the precise location in which a particular cell phone was located –only the phone’s general location relative to the sector of the cell tower with which it was communicating.
III. LEGAL STANDARD A. Federal Rule of Evidence 702 & Daubert
Rule 702 lays out the requirements for a district court to admit expert testimony. The rule
provides that a witness “qualified as an expert by knowledge, skill, experience, training, or
education,” Fed. R. Evid. 702, may offer opinion testimony if it “rests on a reliable foundation and
is relevant to the task at hand,’”
Daubert
,
(a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; Fed. R. Evid. 702. Rule 703 further identifies the permissible “bases” of an expert's opinion field would reasonably rely on those kinds of facts or data in forming an opinion on the subject, aware of or personally observed.” Fed. R. Evid. 703. Additionally, “[i]f experts in the particular testimony, which must be premised on “facts or data in the case that the expert has been made [the facts or data] need not be admissible for the opinion to be admitted.” [8] methods to the facts of the case. (d) the expert has reliably applied the relevant principles and and
The Supreme Court has elaborated the standards for admissibility under Rules 702 and 703
in a series of cases, including ;
General Electric Co. v. Joiner
,
“Expert opinion testimony is relevant if the knowledge underlying it
has a valid connection to the pertinent inquiry. And it is reliable if
the knowledge underlying it has a reliable basis in the knowledge
and experience of the relevant discipline.” [
Primiano v. Cook,
598
26
27
[8]
“[I]f the facts or data would otherwise be inadmissible,” however, “the proponent of the opinion
may disclose them to the jury only if their probative value in helping the jury evaluate the opinion
28
substantially outweighs their prejudicial effect.” Fed. R. Evid. 703.
F.3d 558, 565 (9th Cir. 2010)] (citation and internal quotation marks
omitted). “Shaky but admissible evidence is to be attacked by cross
examination, contrary evidence, and attention to the burden of proof,
not exclusion.”
Id.
at 564 (citation omitted).
The judge is
“supposed to screen the jury from unreliable nonsense opinions, but
not exclude opinions merely because they are impeachable
.”
[
Alaska Rent–A–Car, Inc. v. Avis Budget Grp., Inc.,
at 969–70.
The test of reliability is flexible.
Estate of Barabin v. AstenJohnson,
Inc.,
The proponent of expert testimony “has the burden of establishing that the pertinent
admissibility requirements are met by a preponderance of the evidence.” Fed. R. Evid. 702
advisory committee note—2000 amendment (citing
Bourjaily v. United States
,
B. Federal Rule of Criminal Procedure 16
Rule 16(a) governs the Government’s disclosure obligations in a federal prosecution. Rule 16(a)(1)(F) provides that “the government must permit a defendant to inspect . . . the results or reports of any physical or mental examination and of any scientific test or experiment” where, inter alia , “the item is material to preparing the defense or the government intends to use the item in its case-in-chief at trial.” Fed. R. Crim. P. 16(a)(1)(F). Rule 16(a)(1)(G) addresses disclosures pertaining to a Government’s proposed expert witnesses. It states that “the government must give to the defendant a written summary of any testimony that the government intends to use under Rules 702, 703, or 705 of the Federal Rules of Evidence during its case-in-chief at trial.” Fed. R. Crim. P. 16(a)(1)(G). The summary “must describe the witness’s opinions, the bases and reasons for those opinions, and the witness’s qualifications.” Id. The Government’s disclosure obligations with respect to expert testimony are not
excessively demanding:
Rule 16(a)(1)(G) requires that the government provide a summary of
the opinions of its experts to be used during its case-in-chief. The
rule requires that “[t]he summary . . . must describe the witness's
opinions, the bases and reasons for those opinions, and the witness's
qualifications.” The Advisory Committee Note to the 1993
Amendment stated that the bases and reasons must be sufficient to
allow counsel to frame a motion (or other motion
in
limine
), to prepare for cross-examination, and to allow a possible
counter-expert to meet the purport of the case-in-chief testimony.
United States v. Cerna
,
C. Federal Rule of Evidence 403
Rule 403 provides that a court “may exclude relevant evidence if its probative value is
substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing
the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative
evidence.” Fed. R. Evid. 403. “‘Undue prejudice’ within [this] context means an undue tendency
to suggest decision on an improper basis, commonly, though not necessarily, an emotional one.”
advisory committee note. The Ninth Circuit has stressed that the “danger of [unfair] prejudice
must not merely outweigh the probative value of the evidence, but
substantially
outweigh it.”
United States v. Haischer
,
IV. DISCUSSION
A. Daubert Objections testimony under Rule 702 is whether she must have expert knowledge not only of historical cell presentation. At issue here, Mr. Wendt contends, “is not only discussion of cellphone and cell site data analysis but also of ESPA, the mapping program that she used to produce her PowerPoint mapping and display processes that will display [Ms. Sparano’s] opinions and analysis.” network operation, but also the use of propriet[ar]y, largely undisclosed, law enforcement related [11] As argued by the parties, the primary question regarding the admissibility of Ms. Sparano’s Opening Br. at 1. And Ms. Sparano, according to Mr. Wendt, lacks “the necessary scientific and technical background to explain the actual methodologies and analytical processes involved” in using the mapping software. Id.
Mr. Wendt offers a host of challenges to Ms. Sparano’s proposed testimony, dividing his opening brief into “three main categories” of “nine arguments.” at 2. Yet the latter do not fit neatly into the former, making orderly analysis of the parties’ positions somewhat difficult. At bottom, Mr. Wendt does not take issue with the way that cell phones connect with cell towers, the essential reliability of CDRs, or the general geographic and temporal information that can be gleaned from CDRs. Instead, his focus is on the mapping program and the inferences that Ms. Sparano draws therefrom in her PowerPoint slides. Mr. Wendt’s principal Daubert -specific arguments are: (1) that Ms. Sparano is not qualified to establish the reliability of ESPA, id. at 10- 13; (2) that ESPA is in fact unreliable, largely because the results it has produced here have not been validated through drive tests, id. at 22-27; and (3) that, aside from the ESPA issue, Ms. Sparano’s analysis of the CDRs themselves is unreliable since she failed to account for additional mapping factors beyond those discussed in her declaration, id. at 28-37. This last argument is developed in further detail in the supplemental brief filed by Mr. Ott and Mr. Diaz, which focuses on the asserted importance of “radio frequency theory” to CAST analysis. Suppl. Br. at 5. The Government responds that Mr. “Wendt’s attempt to focus the Court on ESPA is a red herring, and is irrelevant to the Court’s inquiry” under Rule 702, which here concerns only the reliability of Ms. Sparano’s historical cellular analysis. Opp’n Br. at 1. Courts have regularly found such analysis “not only to be reliable, but routine,” the Government argues, and often “without a hearing.” Id. It further asserts that the PowerPoint presentation “is not [Ms. Sparano’s] actual expert opinion, but merely [an] aid in her presentation to the jury of that opinion,” which is contained in her declaration. Id. Moreover, Mr. Wendt’s implicit contention “that any witness who uses any software or technological tool in their work must be an expert in software engineering to testify” would erect an extremely high bar to all manner of expert testimony, which the Government declares “far afield from the purpose of .” See id. And to the extent that the reliability of ESPA is relevant to the Court’s Rule 702 inquiry, the Government notes that it has previously disclosed all of the CDRs on which Ms. Sparano relied to Defendants such that they “can easily verify for themselves the accuracy of the visual representations in her demonstrative exhibit using publicly available software like Google Maps.” Id. To this last point, the Government adds that Defendants have failed to identify “a single point of inaccuracy” in Ms. Sparano’s PowerPoint presentation that would “suggest that the software is unreliable.” Id.
The Court agrees with the substance of the Government’s argument and finds Defendants’ numerous Daubert -based arguments unpersuasive.
1. Qualifications
As a threshold matter under Rule 702, Defendants do not seriously question whether Ms.
Sparano is qualified to testify as an expert on historical cell site data analysis. Mr. Wendt suggests
in passing that, while Ms. Sparano “has received . . . training in cell phone related investigation,”
she lacks the authority of “an expert on cellphone systems, radio wave propagation, or mapping
software.” Opening Br. at 4. In reviewing the contents of Ms. Sparano’s C.V., however, Mr.
Wendt does not contend that she is unqualified to opine on cellular-records analysis.
See id.
at 9.
And while Mr. Ott and Mr. Diaz question “whether Agent Sparano understands . . . the appropriate
scientific standards in this case,” Suppl. Br. at 5, they also decline to challenge her qualifications
as an expert on historical cellular analysis with any specificity. For its part, the Government
emphasizes Ms. Sparano’s many credentials, including “two CAST certification[] trainings” and
instruction in “basic and advanced pin-point cellular survey analysis.” Opp’n Br. at 9 (citing
Sparano Decl. at EXPERT-00002103). The Government also cites district-court decisions that
have approved “other CAST agents who have been similarly qualified” to proffer expert
testimony. at 9-10;
see also United States v. Howard
,
2. Relevance
Defendants likewise fail to meaningfully challenge the relevance of Ms. Sparano’s
proffered testimony. The closest they come is in a two-paragraph section of Mr. Wendt’s brief, in
which he argues that the timing and location of the phone calls that Ms. Sparano addresses in her
PowerPoint slideshow “is relevant only if foundational information can be presented such as to
demonstrate that there would be a basis” for the testimony.
[12]
Opening Br. at 27-28. The
Government readily acknowledges that this is true, and affirms that it “will present evidence, as
typical in these cases, through other witnesses establishing information regarding the target
phones,” namely that they belonged to Defendants at the times in question. Opp’n Br. at 24;
see
also Howard
,
Defendants focus most of their attention on the purported unreliability of Ms. Sparano’s
testimony. Mr. Wendt acknowledges, however, that several courts “have admitted opinion
testimony based on historical cell phone records of varying sorts,” including in this District.
Opening Br. at 6 (citing,
e.g.
,
Cervantes
,
In
United States v. Hill
,
In his trial testimony, [the expert] emphasized that [the defendant’s] cell phone’s use of a cell site did not mean that [the defendant] was right at that tower or at any particular spot near that tower. This disclaimer saves his testimony. Historical cell-site analysis can show with sufficient reliability that a phone was in a general area , especially in a well-populated one. It shows the cell sites with which the person’s cell phone connected, and the science is well understood. . . . A mathematical error rate has not been calculated, but the technique has been subjected to publication and peer criticism, if not peer review. . . . The advantages, drawbacks, confounds [ sic ], and limitations of historical cell-site analysis are well known by experts in the law enforcement and academic communities. . . . . . . Our concern is that the jury may overestimate the quality of the information provided by this analysis. We therefore caution the government not to present historical cell-site evidence without clearly indicating the level of precision—or imprecision—with which that particular evidence pinpoints a person’s location at a given time . The admission of historical cell-site evidence that overpromises on the technique’s precision—or fails to account adequately for its potential flaws—may well be an abuse of discretion. In this case, however, [the expert’s] testimony on both direct and cross-examination made the jury aware not only of the technique’s potential pitfalls, but also of the relative imprecision of the information he gleaned from employing it in this case. at 298-99 (emphasis added). The Seventh Circuit thus concluded that historical cell record
analysis of the kind at issue here is reliable for Daubert purposes when an expert testifies “that a phone was in a general,” rather than precise, “area” at a given time, and makes clear “the relative imprecision of the information” that the methodology provides. [13] See id.
Numerous district courts before and after
Hill
have reached the same essential conclusion
as the Seventh Circuit in that case.
See
,
e.g.
,
United States v. Hitesman
,
Here, Ms. Sparano’s proffered testimony, as described in her declaration, repeatedly clarifies that “historical cell site data analysis is based on the premise that the approximate (not exact) location of a cell phone can be determined.” Sparano Decl. ¶ 4 (emphasis added); see also id. ¶ 5 (stating that service providers sometimes provide law enforcement with data beyond CDRs “to further define the approximate location of a cell phone”) (emphasis added); id. ¶ 10 (stating that CDRs reveal “the approximate location—not the exact location—of a cell phone at the time” a call was received by a cell tower) (emphasis in original). This point is reinforced when the declaration stresses that the “one overriding factor” that determines which cell tower a phone connects to is the strength and quality of the tower’s signal, “regardless of whether that particular cell tower was the closest to the phone at the time the call was generated.” Id. CDRs, in other words, “record the cell tower that provided the best service to the phone” and not necessarily the tower that was nearest in proximity to the phone when it connected with the tower. Id. As a result (and as the Government insisted at the motion hearing), Ms. Sparano is opining only that “a cell phone would connect to the tower with the strongest, clearest signal” such that the phone “would be in the approximate area of that particular tower at the time” a call was made or text message sent. Hearing Tr. at 42. Ms. Sparano’s opinions in her declaration thus adequately contain the caveat required by Hill and numerous other courts that have found CAST-related testimony reliable for Daubert purposes.
Moreover, during the motion hearing the Court repeatedly emphasized to defense counsel
the limited nature of the testimony that Ms. Sparano proposes to offer, observing that her analysis
identifies only the particular cell tower (or sector within that tower) to which a target phone
connected.
See, e.g.
, Hearing Tr. at 24-25, 46-48, 67-68. The Court also expressly stated that it
“inten[ds] to enforce that limit” at trial, restricting Ms. Sparano to opining only on “approximate
location[s]” of cell phones. at 67-68. In fact, the Court explained that its concerns on this
issue were not entirely ameliorated by the Government’s assertion that Ms. Sparano would only
testify to general or approximate areas in which calls could have originated (rather than offering
opinions about, for example, the likely mileage range of a particular cell tower).
See id.
at 42-46.
The Court made clear that if Ms. Sparano wishes to testify about the towers’ coverage areas with
greater specificity than what she has thus far proffered does then the Government would need to
provide additional factual bases to satisfy Rule 702.
See id.
at 64-68. Again, if she were to opine
as to the size of the radius of a particular cell tower’s sector, she would have to provide a specific
and reliable basis therefor. At this point, however, the Court does not consider Ms. Sparano’s
testimony about the general or approximate locations of the target phones problematic for
purposes of
Daubert
.
Defendants offer three principal arguments for why Ms. Sparano’s testimony fails Rule
702. First, Mr. Wendt argues that Ms. Sparano’s decision not to supplement her historical cell site
analysis with additional cellular mapping techniques, such as drive testing, renders it unreliable.
Opening Br. at 2, 25-27. He points out that some recent caselaw has admitted CAST-like
testimony where an expert corroborated his or her initial CDR-based findings through some “type
of double or triple checking.” at 26. In
Morgan
, for example, the court upheld the reliability
of drive testing under , noting that law-enforcement officials “believe it to be a more
accurate method than historical cell-site analysis at approximating a tower’s coverage area.” 292
F. Supp. 3d at 483-84. Mr. Wendt thus argues that, “[a]ccording to the ruling in
Morgan
, drive
testing is the case specific quality assurance mechanism” for CAST testimony. Opening Br. at 12.
But the
Morgan
court’s finding that drive tests can provide a
sufficient
basis for satisfying Rule
702’s reliability requirement does not mean that drive tests are
necessary
for doing so. And as the
Government points out, Mr. “Wendt presents no authority suggesting [that] the absence of drive
testing renders straightforward historical cellular analysis unreliable.” Opp’n Br. at 18. Rather, as
the Court noted above, the
Morgan
court averred that historical cell site analysis, by itself, is
“routinely accept[ed]” as reliable for
Daubert
purposes.
Additionally, Mr. Wendt argues that Ms. Sparano’s testimony is unreliable because she used “limited data” to conduct her analysis. Opening Br. at 28. Mr. Wendt specifically faults Ms. Sparano for declining to validate her analysis with additional “intelligence gathering and analytical tools” offered by the purveyor of ESPA. [14] Id. at 26-27. He also quotes extensively from a law review article that “explains the technical issues that may affect which cell sites are involved in the transmission of cell phone radio waves,” such as “the technical characteristics [1] of cell sites,” “[2] of the antennas on cellular sites,” and “[3] of the phone,” as well as “[4] environmental and geographic factors.” Id. at 33. Mr. Wendt asserts that these “variables . . . can influence the operation of a cell phone network system and influence the attempts to ‘localize’ phone calls,” and complains that Ms. Sparano neglects to account for them. See id. at 34. This is essentially the same argument that Mr. Ott and Mr. Diaz offer in their supplemental brief in support of Mr. Wendt’s motion. They contend, for example, that “a scientist would consider” such factors as “the strength of the transmitting signal, the height of the tower, the angle of the antenna, and the geography of area” when analyzing cell site location information. Suppl. Br. at 5. Ms. Sparano’s failure to do so, they contend, renders her “approximations of a cell tower’s coverage area” unreliable. See id. at 4.
While courts have observed that factors beyond mere location can affect the strength of a
cell tower’s signal and its resulting coverage area,
see United States v. Medley
, 312 F. Supp. 3d
493, 500-01 (D. Md. 2018), Defendants fail to explain how such factors affect the reliability of the
modest opinions that Ms. Sparano offers here. Again, her testimony simply identifies the
particular cell tower (and sector) that a phone actually connected to at a given time, and so
establishes that the phone must have been within the coverage area of that tower (however large or
small it may have been). This historical information, as reflected in the CDRs, is not affected by
the factors that Defendants identify, which instead concern the
extent
of a tower’s coverage area.
Opp’n Br. at 18. So long as Ms. Sparano’s testimony addresses only the
general
or
approximate
location from which a target phone connected to a cell tower and does not opine as to
the precise extent of a cell tower’s coverage area, her failure to incorporate the factors mentioned
by Defendants does not alter the Rule 702 analysis. Defendants are, of course, free to address the
shortcomings of historical cell site analysis at trial. But “the fact that a particular methodology is
subject to limitations . . . is not a bar to admissibility under
Daubert
.” Opp’n Br. at 13;
see also
,
Defendants’ third, and most strenuously asserted, argument for the unreliability of Ms. Sparano’s testimony is that it depends on ESPA’s mapping software and that Ms. Sparano lacks the technical knowledge to understand how ESPA works. Mr. Wendt devotes a great deal of attention to this issue in his brief, and defense counsel did likewise throughout the motion hearing. See, e.g. , Hearing Tr. at 27-31, 75-91. The basic point, however, is straightforward: while Ms. Sparano has training in techniques “specific to CAST and to forensic cell phone investigation,” she “has no background, training, or education in computer science, and apparently no ESPA- specific training that allows her to explain the design, coding, and computer science aspects of ESPA.” Opening Br. at 12. This is particularly problematic, Mr. Wendt argues, because ESPA’s proprietary technology is available only to law enforcement, making its reliability impossible to determine beyond Ms. Sparano’s and the Government’s representations. Id. at 13. Mr. Wendt thus concludes that “[u]nless the Government can establish that Agent Sparano has knowledge of the coding of the algorithms, the operational characteristics of the algorithms, the artificial intelligence related properties of the algorithms, and the methodologies used to verify the accuracy of the resulting maps,” her testimony must be excluded or limited. Id. In sum, as the Government puts it, Mr. “Wendt essentially is objecting that SA Sparano is not an expert on ESPA and that he personally does not have access to the ESPA software.” Opp’n Br. at 14.
The Government responds that the map illustrations that Ms. Sparano seeks to present at
trial are “merely an aid in illustrating her opinion,” rather than the opinion itself.
Id.
Her
testimony addresses the general methodology of historical cell site analysis and “the relevant
CDRs in this case around the time of [a victim’s] murder”; the demonstrative exhibit, in contrast,
simply presents that information “through visual depictions of which towers the phones connected
to.” at 15-16. And creating the exhibit, the Government contends, “requires no expertise
whatsoever” since (as with open-source technology such as Google Maps) it simply involves
“inputting coordinates from CDRs into . . . mapping software,” which “can be done by anyone.”
at 15;
see also United States v. Evans
,
As the Court indicated at the motion hearing, it disagrees with the Government that Ms. Sparano’s slideshow is simply a “visual depiction” of her substantive testimony (as contained in her declaration) and therefore irrelevant to the Rule 702 analysis. See Hearing Tr. at 70-71; see also Wendt Reply at 10 (“The Government’s argument that the slides, maps, and displays are not part of [Ms. Sparano’s] opinions and proposed testimony . . . does not square with reality.”). Insofar as the slides in Ms. Sparano’s PowerPoint presentation convey inferences drawn from the underlying CDRs, they remain subject—independently of the statements in Ms. Sparano’s declaration—to Rule 702’s requirements that her methodology be reliable and that it rest on an adequate factual basis. Fed. R. Evid. 702.
Beyond this clarification, however, the Court agrees with the Government that Ms.
Sparano’s use of the ESPA program does not render her testimony unreliable for Daubert purposes. Defense counsel asserted throughout the motion hearing that firsthand access to ESPA is necessary “to ensure that the software system did, in fact, map each specific transaction correctly.” Hearing Tr. at 79. But as the Government observes, it has long since provided Defendants with “all of the CDRs” underlying Ms. Sparano’s PowerPoint presentation, as well as earlier drafts of the slideshow. See Opp’n Br. at 20. Defendants, however, “have identified no specific areas where the draft presentation is unreliable,” such as by placing a particular cell tower in the wrong location on one of the mapping images. See id. Nor have Defendants ever offered a persuasive response to the question—posed repeatedly throughout the hearing—of why they cannot verify the accuracy of the cell tower locations represented in the slideshow by inputting the CDR data into a publicly available mapping program such as Google Maps. See, e.g. , Hearing Tr. at 29-31, 75-83. The apparent absence of any inaccuracies in Ms. Sparano’s presentation strongly suggests that even if ESPA operates like a “black box,” as Mr. Wendt contends, Defendants have not been harmed by their lack of direct access to the program. Cf. Opening Br. at 24. In sum, given “[1] that the maps produced by ESPA can be independently verified, and [2] the lack of any specific points of unreliability with respect to the maps,” the Court cannot find any reason to exclude or limit Ms. Sparano’s testimony on these grounds. See Opp’n Br. at 20. ESPA, as used by Ms. Sparano here, is simply a mapping function, one that can be performed manually or with publicly available software such as Google Maps, and whose accuracy can be easily verified. Indeed, defense counsel admitted at the hearing that they had no concrete reason to doubt the accuracy of the mapping illustrations in Ms. Sparano’s slideshow or the CDR data on which the slides are based. Hearing Tr. at 36-37 (stating that counsel did not “have an immediate contrary argument to present” about the slides’ presentation of the CDR data); id. at 80 (stating that counsel did not “have a case-specific reason” to doubt the accuracy of the CDRs). In the final analysis, Ms. Sparano’s use of ESPA is not rocket science.
Informing the Court’s conclusion are the sweeping and counterintuitive implications of
Defendants’ position on the ESPA issue. If courts required expert witnesses to possess expert
knowledge of “the software used to generate” demonstrative exhibits such as maps, as Defendants
suggest they should, then law-enforcement and intelligence officials would almost always be
barred from relying on such commonplace exhibits at trial.
See id.
at 16, Hearing Tr. at 16-17.
Similarly, “anyone who testifies using any basic software such as Excel . . . to provide financial
analysis[] would be required to be an expert in the algorithms by which Excel codes its formula
and calculations.” Opp’n Br. at 17. As a result, “no expert utilizing any technological tools would
be permitted to testify without also being an expert software engineer.” at 19. The Federal
Rules of Evidence do not mandate such an absurd result. Indeed, the court in
Morgan
—a decision
that Mr. Wendt cites with approval throughout his briefs,
see, e.g.
, Opening Br. at 11-12, 42-43—
specifically stated that expert witnesses need not possess “an in-depth knowledge of all algorithms
underlying their technological tools . . . to reliably testify about the outputs of those tools.”
See
The Court therefore DENIES Mr. Wendt’s motion to exclude or limit Ms. Sparano’s proffered testimony on the basis of Rule 702 and . The Court emphasizes that, as stated at the motion hearing, Ms. Sparano may not substantially alter or supplement her testimony in any manner that exceeds the bounds that this order has set around her testimony. See Hearing Tr. at 72-73. The Court will set a date by which the Government must submit its final exhibits well in advance of trial. See id.
B. Additional Objections
As Defendants’ arguments in support of the instant motion often merge with one another, see Hearing Tr. at 21, many of the non- -specific issues they raise have been dealt with above. The Court therefore addresses the remaining issues only briefly. 1. Federal Rule of Criminal Procedure 16
Defendants object that the Government has failed to satisfy its disclosure obligations under
Federal Rule of Criminal Procedure 16(a)(1)(G), which requires the Government to provide
defendants with “a written summary” of a proposed expert witness’s “opinions, the bases and
reasons for those opinions, and the witness’s qualifications.” Opening Br. at 14-15, Fed. R.
Crim. P. 16(a)(1)(G).
Mr. Wendt asserts, for example, that Ninth Circuit authority “underscores the need for the
Government to produce foundational information when computer operations and software issues
are central to important aspects of the Government’s case against an accused.”
Id.
at 15. But his
reliance on
United States v. Budziak
,
As indicated above,
Budziak
is inapposite. The precise operation of the contested software
program in that case went directly to the defendant’s degree of criminal intent. Here, in contrast,
the manner in which “ESPA functions does not pertain to a required
element
of the charged
offenses.” Opp’n Br. at 23 (emphasis in original);
see also United States v. Robinson
, 2018 WL
5077260, at *3 (E.D. Mich. Oct. 18, 2018) (rejecting the defendant’s analogy between his request
for access to ESPA and the disclosure issue in
Budziak
, which “went directly to defendant’s
defense of whether he intentionally distributed the pornography or not”). The Government
correctly argues that “[t]he only relevance of ESPA” in the instant case “is that it was used to
create demonstrative maps to aid in SA Sparano’s testimony, which the defense can easily verify
using any publicly available mapping program.”
Id.
The parties’ dispute over CAST-related disclosures, moreover, has been actively litigated
since November 2020, and was effectively resolved at the January 13, 2021, status conference.
There, the Court instructed the Government to make additional disclosures concerning ESPA and
its internal peer review of Ms. Sparano’s analysis.
See
Opp’n Br. at 22. The Government did so
two days later, confirming that Ms. Sparano relied on ESPA in preparing her PowerPoint
slideshow, directing Defendants to the ESPA website for additional information about the
program’s operation and capabilities, and providing a step-by-step explanation of how Ms.
Sparano used ESPA to generate the slideshow’s mapping illustrations. Docket No. 1547-1 at
1. The Government further clarified how CAST’s internal peer review process functions. at 2.
These disclosures easily satisfy both Rule 16 and the Court’s January 13, 2021, discovery order.
See Cerna
,
2. Authentication
Mr. Wendt next contends “that there is no authenticating evidence offered in the
Government’s CAST proffer or in the disclosures related to Agent Sparano’s proposed testimony
that indicate that the Government is undertaking to authenticate Agent Sparano’s mapping
evidence within the meaning of F.R.E. 901(a).” Opening Br. at 17. Under Rule 901(a), the
proponent of an item of evidence “must produce evidence sufficient to support a finding that the
item is what the proponent claims it is.” Fed. R. Evid. 901(a). More specifically, the proponent of
statements produced by machines “must show that a machine is reliable and correctly calibrated,
and that the data put into the machine,” such as “GPS coordinates,” “is accurate.”
United States v.
Lizarraga-Tirado
,
Mr. Wendt does not appear to be concerned with issues of authentication proper, given that
Rule 901(a) generally sets a “low bar” to admissibility, “that SA Sparano was the one who
created” the mapping illustrations, and that she has copious “experience using [ESPA] in her job.”
Opp’n Br. at 25;
see also United States v. Espinal-Almeida
,
3. Confrontation Clause
Mr. Wendt also asserts that Ms. Sparano’s opinions contain testimonial hearsay and thus
run afoul of both the Federal Rules of Evidence and the Confrontation Clause of the Sixth
Amendment to the U.S. Constitution. Mr. Wendt acknowledges that “cell phone records in
themselves if established to be business records can be admitted under F.R.E. 803(6),” an
exception to the general prohibition on hearsay evidence for records that are “kept in the course of
a regularly conducted activity of a business.”
See
Opening Br. at 37, Fed. R. Evid. 803(6). Mr.
Wendt suggests, however, that the exception does not apply to “the records that are prepared by a
cell phone company pursuant to a warrant,”
i.e.
, CDRs.
Id.
But as the Government points out,
numerous courts have held that cell phone records of the type at issue here are business records
within the meaning of Rule 803(6). Opp’n Br. at 25;
see also United States v. Yeley-Davis
,
Mr. Wendt also contends that any “specific entries made by an expert in an illustration or a
map,” such as “tacking . . . a label on a map with a name or coordinates entered,” may amount to
hearsay. Opening Br. at 38. But this argument is foreclosed by
Lizarraga-Tirado
, which held that
while a manually placed and labeled tack on a map is “classic hearsay,” a “tack placed by the
Google Earth program and
automatically
labeled with GPS coordinates isn’t hearsay” because the
tacking is not the assertion of a “person” under Rule 801(a). at 1109-10 (emphasis added).
Here, unlike in
Lizarraga-Tirado
, Ms. Sparano is the testifying witness who made the entries, and
ESPA automatically generates the mapping display.
Mr. Wendt’s further attempts to analogize Ms. Sparano’s testimony to that proffered in
Bullcoming v. New Mexico
,
4. Federal Rule of Evidence 403
Finally, Mr. Wendt raises the issue of whether Ms. Sparano’s PowerPoint presentation is unfairly prejudicial to Defendants under Rule 403. That rule gives the Court discretion to “exclude relevant evidence if its probative value is substantially outweighed by a danger of,” among other things “unfair prejudice,” “misleading the jury,” or “wasting time.” Fed. R. Evid. 403. Mr. Wendt’s essential argument is that “the Government is seeking to introduce to the jury a set of synthesizing illustrations based on an incomplete set of cell phone company records . . . without any of the limiting information or fulsome explanations of the weaknesses and error rates involved in the process used.” Opening Br. at 43. The Government counters that “under Rule 403, the absence of any basic labels on the demonstrative maps will serve to confuse the jury and unnecessarily prolong the testimony.” Opp’n Br. at 27.
Defendants point to a number of proposed slides that, in their view, contain “an argument or an explanation . . . that goes beyond data entry,” either in the form of graphics or text, and thus run afoul of Rule 403. at 38. Several of these slides contain arrows that represent the direction in which target phones purportedly traveled in a given timeframe, based on the sequence of relevant CDR data. See, e.g. , Slides 20, 23, 48, and 49. Defense counsel object to these types of graphics as conveying Ms. Sparano’s interpretation of the underlying data in argumentative fashion rather than neutrally presenting the data. See, e.g. , Hearing Tr. at 57-58. In response, the Government asserts that the graphics visually display permissible (and, in fact, irrefutable) inferences based on the CDR data. at 59.
The Government is correct that the slides’ representations of simple deductions about a phone’s direction of travel based on the timing and sequence of general location data are not complicated and may be highly probative. See id. at 52, 59-60. The representations are also not unduly prejudicial since any errors or shortcomings will be subject to cross-examination. The Court is more receptive, however, to defense counsel’s argument that an express statement in slide 39 that two Defendants were “traveling together” is problematic insofar as it suggests a degree of concerted action that may not be supported by the underlying data. See id. at 52-55. In this respect, the statement could be excludable either as lacking an adequate factual basis for Daubert purposes or as likely to mislead the jury under Rule 403.
The parties shall use the guidance herein to resolve further evidentiary disputes over Ms. Sparano’s proposed testimony. The Court expects such disputes to be resolved before trial via objections lodged when exhibits are disclosed prior to the pretrial conference.
V. CONCLUSION For the reasons given above, the Court DENIES Mr. Wendt’s motion to exclude or limit Ms. Sparano’s proffered testimony on the grounds that it fails to satisfy Federal Rule of Evidence 702 and . The Court also DENIES Mr. Wendt’s motion on the alternative theories he has asserted thus far. The Court denies Defendants’ additional objections to Ms. Sparano’s demonstrative exhibit based on Federal Rule of Evidence 403, except as provided herein.
This order disposes of Docket No. 1444. IT IS SO ORDERED .
Dated: April 13, 2021 ______________________________________ EDWARD M. CHEN United States District Judge
Notes
[1] Judge Beeler nevertheless requested (but did not order) that the Government provide Defendants 23 with a copy of “the actual PowerPoint” presentation that Ms. Sparano proposes to use at trial, rather than the PDF version of it that the Government apparently turned over to Defendants on or 24 around December 18, 2020. Id. ; see also Docket No. 1511 (“Reply Br.”) at 1 n.1 and 5. n.5 (describing Defendants’ receipt of the slideshow dated December 18, 2020); Opening Br. at 8-9 25 (describing earlier versions of the slideshow that Defendants have received). 26
[2] As Government’s discovery letter explains, ESPA “stands for Enterprise Sensor Processing Analytics.” Docket No. 1547-1 at 1. The proprietary software “is a product of Gladiator 27 Forensics.” Id. The Government represents that “CAST agents like SA Sparano receive virtual online training as well as in-person, hands-on training from Gladiator Forensics.” 28
[3] The PowerPoint slideshow, Docket No. 1490-2, has been filed under seal. 25
[4] CAST agents, Sparano explains, “provide cellular record analysis and technological assistance to 26 local, state, and federal law enforcement, including mapping, analytical report[s], and expert testimony at trials.” Sparano Decl. ¶ 1. 27
[5] The Government also includes a copy of Sparano’s two-page C.V. with her declaration. See 28 Docket No. 1490-1 at 8-9.
[6] Mr. Sparano further explains, in her PowerPoint presentation, that cell tower sectors generally 24 “cover approximately 120” degrees each, so that each tower is comprised of three sectors. See Slides at 4. The angle of the open wedge graphics in her slideshow reflects this 120-degree sector 25 size. The PowerPoint presentation also clarifies that the shaded area within each open wedge graphic “shows the general direction the radio frequencies (RF) travel from the antennas “and 26 “does not depict the actual . . . coverage area of the particular sector.” at 5 (emphasis in original). 27
[7] Alternative, less direct, methods of verifying the accuracy of cell site location information 28 include “video surveillance, license plate readers, defendant’s admission, witness statements,
[9] The Ninth Circuit recently overruled
Estate of Barabin
on other grounds in
United States v.
28
Bacon
,
[10] As previously noted, Defendants also raise objections to Ms. Sparano’s proffered testimony 24 based on other grounds, including Federal Rules of Evidence 901 and the Confrontation Clause. The standards governing those issues will be addressed briefly below. 25
[11] See also Opening Br. at 6 (“The issues presented here have to do not only with what analytical 26 technology has been disclosed by the Government, but also with the vagaries and uncertainties about the reliability of illustrative slides, charts, PowerPoints, and maps that are sought to be 27 introduced by the Government.”); Reply Br. at 1 (“[T]he concern here is not the basic theory of radio transmission—it is the methodology used to arrive at the case specific opinions, and 28 arguments, that are proposed to be displayed to the jury.”).
[12] Mr. Wendt frames this objection as presenting a Rule 403, rather than a , issue. See 28 Opening Br. at 28.
[13] At the motion hearing, counsel for Mr. Wendt acknowledged that “the state of the law” on the admissibility of cellular record analysis under Rule 702 “is with respect to what’s said in Hill ,” a case that is “cited fairly often.” Hearing Tr. at 20.
[14] Mr. Wendt specifically faults Ms. Sparano for declining to validate her analysis with additional “intelligence gathering and analytical tools” offered by the purveyor of ESPA. Id. at 26-27. He also points to the absence of AT&T Network Event Location Services (“NELOS”) data in Ms. Sparano’s analysis, even though NELOS records purportedly provide a more accurate estimate of a cell phone’s location at a given time than historical cell site information. at 29. He therefore accuses Sparano and the Government of using “only some of the available information that AT&T provides when litigants obtain AT&T records.” at 30.