THE PEOPLE, Plaintiff and Respondent, v. DANIEL MEZA et al., Defendants and Appellants.
B318310
IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA SECOND APPELLATE DISTRICT DIVISION SEVEN
Filed 4/13/23
CERTIFIED FOR PUBLICATION; (Los Angeles County Super. Ct. No. TA150314)
Sharon Fleming, under appointment by the court of appeal, for Defendant and Appellant Daniel Meza.
Bess Stiffelman, under appointment by the court of appeal, for Defendant and Appellant Walter Meneses.
Rob Bonta, Attorney General, Lance E. Winters, Chief Assistant Attorney General, Susan Sullivan Pithey, Senior Assistant Attorney General, Wyatt E. Bloomfield and Michael C. Keller, Deputy Attorneys General, for Plaintiff and Respondent.
Jennifer Lynch and Andrew Crocker for Electronic Frontier Foundation as Amicus Curiae on behalf of Defendants and Appellants.
INTRODUCTION
“A geofence is a virtual fence or perimeter around a physical location. Like a real fence, a geofence creates a separation between that location and the area around it. . . . [¶] It can be any size or shape, even a straight line between two points. [¶] Geofences are created using mapping software, which allow the user to draw the geofence over the desired geographic area. It is made up of a collection of coordinates (i.e., latitude and longitude) or in the case of a circular geofence one point that forms the center.”1
“Geofence warrants (sometimes called ‘reverse location searches‘) are official requests by law enforcement authorities to access the device location data gathered by large tech companies like Google. The warrants specify a time and geographic area, and require the companies to turn over information
“The government filed its first geofence search warrant in 2016, and by the end of 2019, Google was receiving about 180 search warrant requests per week from law enforcement officials across the country. . . . Between 2018 and 2020, Google received about 20,000 geofence warrant requests for data, including over 11,500 in 2020 alone.”3
* * *
Daniel Meza and Walter Meneses were identified as suspects in the murder of Adbadalla Thabet after a geofence search warrant directed to Google revealed cell phones signed in to Google accounts connected to them were in several of the same locations as Thabet on the day of his murder. After their motions to quash and suppress evidence were denied, Meza pleaded guilty to first degree murder; and Meneses pleaded no contest to second degree murder.
On appeal Meza and Meneses contend the trial court erred in denying their motion to suppress, arguing the geofence warrant violated their rights under the
1. The Murder of Adbadalla Thabet and the Initial Investigation
According to surveillance footage viewed by police officers, at approximately 10:30 a.m. on March 1, 2019 Thabet drove into the parking lot of a bank in Paramount, followed by a gray sedan and a red sedan.5 The driver of the red car parked, got out of his vehicle and walked to the gray car, where he stopped to speak to the driver of the gray car. The driver of the gray car then drove slowly toward Thabet‘s parked car. The driver of the red car followed on foot. As Thabet got out of his vehicle, the gray car pulled up next to Thabet‘s car; and an occupant of the gray car shot Thabet in the torso. Thabet fell to the ground as the gray car sped away. The driver of the red car approached Thabet, took his backpack, retreated to the red car and drove away. Thabet died from his injuries.
The investigating officers were able to retrace Thabet‘s steps from the morning of the shooting. They learned Thabet worked for his uncle‘s business, which included managing several gas stations. Twice per week Thabet picked up cash receipts from the gas stations and deposited the cash at the bank in Paramount. The day of the shooting Thabet left his apartment building in Downey around 7:00 a.m. and drove to a gas station in Downey to pick up cash for deposit. Thabet was at the Downey gas station from approximately 7:15 a.m. to 7:30 a.m. Thabet then met his brother-in-law at approximately 9:00 a.m. at a gas station in Bellflower. Thabet and his brother-in-law departed the gas station in separate cars at approximately 9:40 a.m. and drove to a strip mall in Compton where the brother-in-law was contemplating renting retail space. Thabet left the strip mall alone, driving to a gas station in Lynwood to pick up cash receipts. From Lynwood Thabet drove to the bank in Paramount where he was killed.
In addition to the video surveillance from the bank parking lot, investigators obtained video surveillance from other locations Thabet visited that morning. The gray and red vehicles from the bank surveillance footage were also identified in surveillance footage from at least two of those additional locations. Investigators concluded the suspects had been following Thabet, anticipating his arrival at the bank with the cash deposits. The license plate numbers of the gray and red vehicles were not legible in any of the footage.
a. Probable cause
Los Angeles County Sheriff‘s Detective Jonathan Bailey applied for a search warrant directing Google to identify individuals whose location history data indicated they were in the vicinity of the six locations visited by Thabet on March 1, 2019. In an affidavit supporting the application, Bailey described Thabet‘s murder as seen on the surveillance footage of the bank parking lot. Bailey stated he had viewed surveillance camera footage from several of the other locations Thabet had visited that morning and had seen the gray and red sedans in the footage. Bailey did not state how many of the six locations had available surveillance footage, nor did he identify the locations at which the red and gray cars were visible.
The affidavit included a brief overview of how Google tracks and stores location history data, stating Google collected data through “Global Position System (GPS) data, cell site/cell tower information, Bluetooth connections, and Wi-Fi access points.” Bailey stated, “I know most people in today‘s society possess cellular phones and other items (e.g. tablets, watches, laptops) used to communicate electronically. . . . Most people carry cellular phones on their person and will carry them whenever they leave their place of residence.” In addition, Bailey explained, “Suspects involved in criminal activity will typically use cellular phones to communicate when multiple suspects are involved.” Therefore, Bailey concluded, identification of individuals in Thabet‘s vicinity on the day of the murder would assist investigators in locating the drivers of the vehicles involved in the murder, who investigators believed had been following Thabet throughout the morning.
b. Search parameters
The warrant application sought location history data for individuals within six target locations. The first location was Thabet‘s apartment, which was located in the middle of a large city block, surrounded by both residential and retail buildings. The area designated for the search was a circle with a radius of 100 meters from the center of the apartment complex (approximately seven and a half acres). It included the entire apartment building as well as portions of several surrounding buildings and approximately three-quarters of the street in front of the building. The timeframe for this search was 6:00 a.m. to 7:15 a.m. on March 1, 2019.
The second location was the gas station in Downey where Thabet picked up cash for deposit. The gas station is on the corner of a large intersection and is surrounded by other retail establishments. The search area consisted of
The third location was the gas station in Bellflower. The search area consisted of a circle with a radius of 50 meters from the approximate center of the gas station (almost two acres). Included in the circle were part of the intersection and approximately 50 meters of the streets bordering the gas station, as well as portions of the surrounding businesses. The timeframe for this search was 7:30 a.m. to 9:40 a.m.
The fourth location was the strip mall in Compton. The search area was a rectangle that included the strip mall, three streets bordering it and some neighboring buildings and parking lots (approximately one and one-half acres). The timeframe for this search was 9:40 a.m. to 10:15 a.m.
The fifth location was the gas station in Lynwood. The search area was a rectangle that included the gas station, neighboring buildings, including buildings across the street that appeared to be residences and the intersection bordering the gas station (approximately three acres). The timeframe for this search was 10:15 a.m. to 10:30 a.m.
The sixth and final location was the bank in Paramount where the murder took place. The search area was a circle with a radius of 75 meters from the center of the bank building (more than four acres). The search area included the bank and parking lot, neighboring businesses and parking lots, the intersection in front of the bank and approximately 50 meters of the streets bordering the bank.
c. The warrant process
The warrant set forth a three-step process by which Google would respond to the request for information. At step one, Google was directed to search location history data for the six designated locations and times and produce an anonymized list of devices found within the search areas in the designated timeframes, including the individual times each device was recorded in the search area during the applicable time period.
At step two, law enforcement would review the anonymized list of devices “to remove devices that are not relevant to the investigation, for example, devices that were not in the location for a sufficient period of time.” If law enforcement believed additional information was needed to determine
At step three, law enforcement could demand identifying information from Google for all devices law enforcement deemed relevant to the investigation. The warrant directed Google to provide this identifying information without additional legal process.
3. Execution of the Search Warrant and Charges Against Meza and Meneses
A Los Angeles superior court judge, acting as magistrate, signed the geofence search warrant on March 21, 2019.
After reviewing the anonymized data provided by Google, the Sheriff‘s Department sought identifying information for eight devices that had been in the relevant locations on March 1, 2019. Google provided corresponding email addresses to law enforcement. The Sheriff‘s Department then drafted additional search warrants related to two of those email addresses, which eventually led to the identification of Meza and Meneses as suspects.
In an information filed December 4, 2020 Meza and Meneses were charged with murder (
4. The Motion To Quash and Suppress
On March 18, 2021 Meza moved pursuant to
A hearing on the motion was held on April 12, 2021. Spencer McInvaille, an expert on geolocation and mobile devices, testified on behalf of Meza and Meneses. McInvaille‘s testimony was based on his training and experience, as well as his review of documents publicly filed by Google.6
McInvaille explained Google‘s location data is derived from several sources: GPS, Bluetooth signals, cellular network data and the strength of nearby WiFi networks.7 Google logs each device‘s location hundreds of times each day—as often as every two minutes according to some estimates. (See United States v. Chatrie (E.D.Va. 2022) 590 F.Supp.3d 901, 908 & fn. 10 (Chatrie).) However, Google cannot pinpoint a user‘s location with 100 percent accuracy. McInvaille stated the longitude and latitude recorded by Google as the device‘s location is “not a physical actual location of the device. It‘s just the estimate derived from the measurement that they took.” Thus, Google also reports a confidence interval, measured in meters, that indicates Google‘s confidence in the location of the device. For example, a confidence interval of 15 meters indicates Google estimates the device is within a 15 meter radius of the given coordinates. The size of the confidence interval varies depending on the type of data from which the measurement was taken.
Google aims to estimate a device‘s location with 68 percent accuracy—that is, there will be a 68 percent chance the user was actually within the circle created by the confidence interval. When responding to a geofence warrant, Google considers a device within the search parameter if the estimated location is within the search boundaries even if the confidence interval extends beyond the search
Romy Haas, a crime analyst for the Sheriff‘s Department, testified for the prosecution regarding the application for and execution of the geofence warrant in this case. Haas explained she typically consults with detectives prior to drafting a geofence warrant application and assists in establishing the geographic parameters and timeframes of the requested warrant. She had participated in drafting and processing returns on more than 50 geofence warrants by the time of the motion to suppress hearing in 2021, but at the time she assisted Detective Bailey with drafting the geofence warrant in this case in 2019 she had worked on only two other geofence warrants. Haas had participated in a number of trainings regarding location history data and geofence warrants, most of which took place after the warrant had been drafted in this case.
The court directly questioned Haas regarding how she and Detective Bailey decided on the search parameters for the warrant. For the first location (Thabet‘s apartment building), Haas testified the search radius of 100 meters from the center of the apartment building was selected so that it would capture the street in front of the building “in the event that [Thabet] was being watched before he left.” Haas explained it was typical with early geofence warrants to draw a circle from a midpoint, but she noted a polygon “will help reduce the number of devices that will show up in the geofence.”
For the second location, Haas testified the geofence perimeter was again drawn to capture the streets bordering the gas station “to see if . . . someone had been coming down those streets or parked on those streets if the—if someone was watching the victim at that location.” The perimeter for the third location was drawn to include the street on the north side of the gas station because there was surveillance video footage showing the suspect vehicles parked on that street. The perimeters for locations four and five were drawn as rectangles because Haas found using a circle captured too much area and would “encompass a bunch of devices that I didn‘t feel would be necessary because they were in the outer neighborhood.” Instead, for location four she drew a rectangle that encompassed the area the victim visited inside the strip mall, and for location five the rectangle encompassed a parking lot across the street from the gas station where one of the suspect vehicles had been seen on surveillance video.
Haas also testified regarding the warrant‘s three-step process for Google‘s production of data in response to the warrant. Haas explained the process was mandated by Google as the procedure that would most likely ensure Google‘s
Google produced a list of eight anonymized accounts that had been at two or more of the six locations at the relevant time periods. Of the eight accounts, one had been at four of the geofence locations, one at three locations and the remaining six had been at two locations. Haas requested, and Google produced, identification information for all eight accounts. Two of those accounts (the ones that had been at three and four of the locations) ultimately led authorities to Meza and Meneses.
5. The Superior Court‘s Denial of the Motions To Suppress
The superior court found there was sufficient probable cause to support issuance of the geofence warrant. The fact that the two suspect cars were seen in multiple surveillance videos made it reasonably probable “that they were using their phones to communicate or to determine the location that they‘re going to.” The court further found the warrant satisfied the particularity requirements of the United States Constitution and CalECPA. The court stated it was satisfied the boundaries of the search areas were based on the locations of the suspect vehicles as seen in the video footage and were not so broad as to unnecessarily include devices of uninvolved bystanders. Finally, the court ruled, even if the warrant had been defective, the officers were entitled to rely on it under the good faith exception of United States v. Leon (1984) 468 U.S. 897 (Leon). Accordingly, the court denied the motions to suppress.
Following denial of the motions to suppress evidence, Meza pleaded guilty to first degree murder and Meneses pleaded no contest to second degree murder. Pursuant to negotiated agreements the special circumstances and special allegations were stricken, and the remaining counts as to Meneses were dismissed. Meza was sentenced to an indeterminate state prison term of 25 years to life. Meneses was sentenced to an indeterminate state prison term of 15 years to life.
DISCUSSION
1. The Geofence Warrant Violated the Fourth Amendment
a. Governing Law and Standard of Review
The
A search is presumptively reasonable, and thus in compliance with the
Accordingly, in determining the validity of a warrant, courts examine three main factors: probable cause, particularity and overbreadth.10 Probable cause will be found to support the issuance of a warrant if “the magistrate had a substantial basis for concluding a fair probability existed that a search would uncover wrongdoing.” (People v. Westerfield (2019) 6 Cal.5th 632, 659-660; accord, People v. Miles (2020) 9 Cal.5th 513, 576; People v. Kraft (2000) 23 Cal.4th 978, 1040-1041; see Illinois v. Gates (1983) 462 U.S. 213, 238-239 (Gates).)
“Particularity is the requirement that the warrant must clearly state what is sought.” (In re Grand Jury Subpoenas Dated Dec. 10, 1987 (9th Cir. 1991) 926 F.2d 847, 856.) To satisfy this requirement, “[c]omplete precision in describing the place to be searched is not required.” (People v. Amador, supra, 24 Cal.4th at p. 392; accord, People v. Minder (1996) 46 Cal.App.4th 1784, 1788.) “It is enough if the description is such that the officer with a search warrant can with reasonable effort ascertain and identify the place intended.” (Amador, at p. 392; accord, Steele v. United States (1925) 267 U.S. 498, 503.)
“Breadth deals with the requirement that the scope of the warrant be limited by the probable cause on which the warrant is based.” (In re Grand Jury Subpoenas Dated Dec. 10, 1987, supra, 926 F.2d at pp. 856-857.) This
“In reviewing the trial court‘s suppression ruling, we defer to its factual findings if supported by substantial evidence. We independently assess the legal question of whether the challenged search or seizure satisfies the Fourth Amendment.” (People v. Brown (2015) 61 Cal.4th 968, 975; accord, People v. Eubanks (2011) 53 Cal.4th 110, 133.)
b. The Search Warrant Was Supported by Probable Cause
Meza and Meneses contend Detective Bailey‘s assertion of probable cause in his affidavit was insufficient because “[t]here was absolutely no evidence that either suspect had, or was using, a phone or other device at any time during the relevant timeframe.” Accordingly, they argue, there was no basis for determining that searching cell phone location history would lead to the identity of potential suspects or the recovery of other evidence related to the murder.
Probable cause does not require conclusive evidence that a search will uncover relevant evidence, only that “there is a fair probability that contraband or evidence of a crime will be found in a particular place.” (People v. Kraft, supra, 23 Cal.4th at p. 1041; accord, Gates, supra, 462 U.S. at p. 238.) “‘[S]ufficient probability, not certainty, is the touchstone of reasonableness under the Fourth Amendment.‘” (People v. Beck and Cruz (2019) 8 Cal.5th 548, 592; see also People v. Carrington (2009) 47 Cal.4th 145, 163 [“[t]he showing required in order to establish probable cause is less than a preponderance of the evidence or even a prima facie case“].) In making this determination a magistrate may draw reasonable inferences about where evidence is likely to be found based on the nature of the evidence and the type of offense. (See Gates, at p. 240; People v. Sandlin (1991) 230 Cal.App.3d 1310, 1315.)
It was reasonable for the magistrate to conclude the perpetrators were carrying cell phones the morning of the murder and used them in coordinating their movements. Not only did Detective Bailey opine, based on his
c. The Search Warrant Lacked Sufficient Particularity
As discussed, the “purpose of the ‘particularity’ requirement of the
The warrant in this case sufficiently described the place to be searched (Google‘s database of users’ location history) and the items to be retrieved from that search (designated records for users found within the boundaries of
However, the warrant here failed to meet the particularity requirement because it provided law enforcement with unbridled discretion regarding whether or how to narrow the initial list of users identified by Google. Once the step one search had been conducted, law enforcement officials were able to enlarge the geographic parameters of the search and request additional information on any of the potentially thousands of users identified without any objective criteria limiting their discretion. Again, at step three law enforcement could seek identifying information of any of the users found within the search parameters without restriction on how many users could be identified or any further showing that information concerning each individual user would be relevant to the case.
This failure to place any meaningful restriction on the discretion of law enforcement officers to determine which accounts would be subject to further scrutiny or deanonymization renders the warrant invalid. (See Chatrie, supra, 590 F.Supp.3d at p. 934 [geofence warrant lacks requisite particularity because “Steps 2 and 3 of this warrant leave the executing officer with unbridled discretion and lack any semblance of objective criteria to guide how officers would narrow the lists of users“]; In re Search of: Info. Stored at Premises Controlled by Google (N.D.Ill. 2020) 481 F.Supp.3d 730, 754 (In re Google N.D.Ill) [denying geofence warrant application because “the warrant puts no limit on the government‘s discretion to select the device IDs from which it may then derive identifying information from among the anonymized list of Google-connected devices that traversed the geofences“]; In re Search of Info. Stored at the Premises Controlled by Google (Va.Cir.Ct., Feb. 24, 2022, KM-2022-79) 2022 Va.Cir. Lexis 12, at pp. *24-*25 [denying geofence warrant application that allowed police to “unilaterally . . . enlarge the Court-authorized search zone” and “unilaterally tell Google which cell phones it wants to unmask to obtain the owner‘s personal information“].)11
In determining whether a warrant is overbroad courts consider “whether probable cause existed to seize all items of a category described in the warrant” and “whether the government could have described the items more particularly in light of the information available to it at the time the warrant issued.” (United States v. Shi (9th Cir. 2008) 525 F.3d 709, 731-732; see also People v. Hepner (1994) 21 Cal.App.4th 761, 778 [“overbreadth also hinges on whether a more precise description [of the items to be seized] was reasonably possible“]; People v. MacAvoy (1984) 162 Cal.App.3d 746, 754-755 [“[o]n its face, the warrant would allow the officers to search every part of the fraternity house; since probable cause existed to search appellant‘s room only, the warrant, as a general rule, is void“]; Owens v. Lott (4th Cir. 2004) 372 F.3d 267, 276 [warrant authorizing search of “all persons” at certain location was valid “if the affidavit and information provided to the magistrate supply enough detailed information to establish probable cause to believe that all persons on the premises at the time of the search are involved in the criminal activity“]; In re Grand Jury Subpoenas Dated Dec. 10, 1987, supra, 926 F.2d at p. 857 [“the concept of breadth may be defined as the requirement that there be probable cause to seize the particular thing named in the warrant“].)
The geofence warrant in this case ran afoul of both of these requirements. First, the warrant authorized the identification of any individual within six large search areas without any particularized probable cause as to each person or their location. For example, the first search location, the area around Thabet‘s apartment complex, allowed law enforcement to obtain information on every individual in a seven-and-a-half-acre area over a 75 minute period in the early morning. The search area included Thabet‘s entire apartment complex and surrounding buildings despite the lack of any evidence (or supported inference) that the suspects left their vehicles, let alone entered the apartment building. Given the early morning timeframe for the search, the warrant permitted identification of numerous individuals with no connection to the murder who were simply still at home. Indeed, for many of the search locations, the geographic boundaries incorporated more surface area where the suspects were not believed to have been present (inside buildings) than area where they were (adjacent roads and intersections). This overbreadth is even more pernicious given that individuals (especially those near the perimeters of the search area) would be included in the warrant return despite an estimated 32 percent chance they were actually not within the search parameter at all.
Second, law enforcement officials failed to draw the search boundaries as narrowly as they could have given the information available. For the first
The timeframes designated in the geofence warrant were also not narrowly tailored. The most striking example of this overbreadth was with location three, the Bellflower gas station where Thabet met his brother-in-law. According to preliminary hearing testimony, Thabet‘s brother-in-law told police he met Thabet at the gas station at approximately 9:00 a.m. and they left at approximately 9:40 a.m. The warrant, however, directed Google to search the location for any devices present between 7:30 a.m. and 9:40 a.m. Even allowing for some uncertainty, there is no evidence Thabet or the suspects were at the gas station 90 minutes before the time that the brother-in-law recalled arriving. Given this was a gas station in a metropolitan area during normal commuting hours, there were likely many devices travelling through the search area during that 90 minutes that were entirely unrelated to Thabet‘s murder.
Haas‘s testimony there was surveillance footage from the Bellflower location showing one or both suspect cars parked on the street near the gas station constituted further evidence of the failure to narrow the parameters. The Sheriff‘s Department presumably could have determined a far shorter time period during which the suspects were present based on a timestamp in the surveillance footage, but they failed to narrow the search accordingly. In fact, the evidence presented to the magistrate was devoid of any detail regarding the surveillance footage that would have supported a finding of probable cause for the particular search areas and times. Detective Bailey‘s affidavit stated only that surveillance footage was available at “several locations” without identifying which locations had surveillance footage and which footage showed the suspects’ vehicles, let alone the precise location and time the suspects’ vehicles were seen. This information should have been used to more narrowly focus the search parameters.
The failure to sufficiently narrow the search parameters potentially allowed a location-specific identification of thousands of individuals—likely a
Other cases that have considered the validity of geofence warrants have also, almost uniformly, determined that such warrants are valid only if they are narrowly tailored to avoid unnecessary infringement on the privacy of uninvolved third parties. For example, in Chatrie a geofence warrant was issued directing Google to search a 17.5-acre area surrounding a bank where a robbery had occurred. The timeframe was for approximately 30 minutes prior to the robbery and 30 minutes after the robbery—a total of one hour. The district court noted the search area included a church and the search identified individuals “who may not have been remotely close enough to the Bank to participate in or witness the robbery.” (Chatrie, supra, 590 F.Supp.3d at p. 930.) The court found the warrant was overbroad because it failed to “include any facts to establish probable cause to collect such broad and
An example at the other end of the spectrum is In re Google D.D.C., supra, 579 F.Supp.3d 62. In that case, police were investigating criminal activity at a business located in an industrial area. Police obtained surveillance footage from inside the business showing the suspects engaging in criminal activity. Based on the precise locations of the suspects and the times depicted in the footage, police designated a geofence area of less than a quarter of an acre, including the front-half of the business and the parking lot but excluding another business in the building and the road bordering the building. The time period in the warrant totaled 185 minutes in increments of two to 27 minutes on 8 different days based on when police knew the suspects had been present. The warrant affidavit also explained that, during the designated time periods, the suspects were either alone inside the business or were in the proximity of “‘on average’ no more than 2 or 3 others.” (Id. at p. 73.) The magistrate judge granted the warrant application, finding the government had “appropriately contoured the temporal and geographic windows in which it is seeking location data” and the warrant did not “have the potential of sweeping up the location data of a substantial number of uninvolved persons.” (Id. at pp. 80 & 85; see also In re Arson Investigation, supra, 497 F.Supp.3d at p. 353 [granting geofence warrant application where search area excluded residences and commercial buildings, time periods sought were approximately 15 to 30 minutes per location and there was evidence premises in search areas were unoccupied during relevant time periods; “the government has structured the geofence zones to minimize the potential for capturing location data for uninvolved individuals and maximize the potential for capturing location data for suspects and witnesses“]; United States v. Rhine (D.D.C. Jan. 24, 2023, No. 21-0687) 2023 U.S.Dist. Lexis 12308, at pp. *95-*103 [denying motion to suppress where geofence warrant had sought identification of individuals in the United States Capitol Building over a four and a half hour period on January 6, 2021; court found warrant was narrowly tailored to include individuals improperly inside the Capitol given that the building was closed
2. The Officers Reasonably Relied on the Geofence Warrant in Good Faith
“In Leon, the [United States] Supreme Court held that when ‘an officer acting with objective good faith has obtained a search warrant from a judge or magistrate and acted within its scope,’ the ‘marginal or nonexistent benefits’ produced by suppressing the evidence obtained ‘cannot justify the substantial costs of exclusion.“’ (People v. Lazarus (2015) 238 Cal.App.4th 734, 766, quoting Leon, supra, 468 U.S. at pp. 920-922.) Accordingly, denial of the motion to suppress must be upheld under the “good faith” exception to the exclusionary rule where a search has been conducted “in objectively reasonable reliance on a subsequently invalidated search warrant.” (Leon, at p. 922.) Leon set forth four scenarios in which such objectively reasonable reliance should not be found and suppression remained the appropriate remedy: (1) “[T]he magistrate or judge in issuing a warrant was misled by information in an affidavit that the affiant knew was false or would have known was false except for his reckless disregard of the truth“; (2) if “the issuing magistrate wholly abandoned his [or her] judicial role“; (3) the affidavit is “so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable“; or (4) if the warrant was “so facially deficient—i.e., in failing to particularize the place to be searched or the things to be seized—that the executing officers cannot reasonably presume it to be valid.” (Id. at p. 923.) The government bears the burden to establish applicability of the good faith exception. (People v. Willis (2002) 28 Cal.4th 22, 36-37.)
Meza and Meneses argue both the third and fourth Leon scenarios—a total lack of probable cause and an obvious failure to satisfy the requirement of particularity—apply here. As discussed, probable cause supported issuance of the warrant. This factor does not preclude application of the good faith exception.
In determining whether the warrant was so facially deficient that the executing officers could not have reasonably presumed it to be valid, “we apply the objective test of ‘whether a reasonably well trained officer would have known that the search was illegal despite the magistrate‘s authorization.“’ (People v. Hochanadel (2009) 176 Cal.App.4th 997, 1015; see also People v. Camarella (1991) 54 Cal.3d 592, 605.) “This objective standard ‘requires officers to
Meza and Meneses argue the good faith exception should not apply here because, instead of following the three steps described in the warrant, “Haas and Bailey disregarded the express terms set forth in the warrant, and essentially fashioned their own search warrant.” While officers may not rely on the good faith exception when they have knowingly exceeded the scope of a warrant (see Leon, supra, 468 U.S. at p. 918, fn. 19 [the good faith exception “assumes, of course, that the officers properly executed the warrant and searched only those places and for those objects that it was reasonable to believe were covered by the warrant“]; see also People v. Nguyen (2017) 12 Cal.App.5th 574, 586-587), the manner of execution in this case (Google‘s filtering of the results at step one) narrowed, not expanded, the search authorized by the warrant. Rather than receiving a list of many thousands of anonymized devices from Google that could then be filtered and matched, law enforcement received only information about eight specific devices—as if only two rooms of a house were searched pursuant to a warrant that authorized searching the entire property.
Given the dearth of authority directly on point and the novelty of the particular surveillance technique at issue, the officers were not objectively unreasonable in believing the warrant was valid, even if the issue, upon close legal examination, is not a particularly close one. (See People v. Rowland (2022) 82 Cal.App.5th 1099, 1124 [applying good faith exception where no California precedent existed on the issue]; People v. Pressey (2002) 102 Cal.App.4th 1178, 1191 [same]; see also United States v. Smith (N.D.Miss. Feb. 10, 2023, No. 3:21-cr-107-SA) 2023 U.S.Dist. Lexis 22944, at pp. *37-*38 [applying good faith exception to geofence warrant given lack of legal authority on the issue]; Chatrie, supra, 590 F.Supp.3d at p. 938 [same].)14
3. The Geofence Warrant Did Not Violate CalECPA
Effective January 1, 2016, CalECPA requires law enforcement officials to obtain a warrant in order to compel production of electronic communication information and electronic device information from a service provider.15 (
Meza and Meneses first argue the geofence warrant in this case violated CalECPA because it “fails to specifically target individuals or accounts. No individual‘s name was included in the warrant, nor was any specific cell phone number, email address, or account information.” Their argument ignores the plain language of the statute, which provides that a warrant shall describe with particularity the information to be seized “as appropriate and reasonable.” (
Meza and Meneses next argue the warrant violated CalECPA because it did not specify the “applications and services covered” by the warrant. CalECPA does not define “applications and services“; and Meza and Meneses have not explained what they believe it means, what particular information they contend should have been included in the warrant, or how the warrant was ambiguous absent such unspecified information. The common sense meaning of the statute appears to be that, when law enforcement seeks to recover the content of electronic communications, such as emails or text messages, the warrant must specify, as appropriate and reasonable, the particular mail or text message applications and services from which law enforcement seeks to retrieve information. With a geofence warrant, however, the government is not seeking data or content related to a particular application or service. Rather, what is sought is the service provider‘s record of all electronic contact with that device, regardless of which applications or services originated the contact. Accordingly, the failure to name a particular application or service in this instance does not result in a violation of CalECPA.
Finally, Meza and Meneses argue any constitutional infirmities in the warrant create an independent violation of CalECPA.17 Meza and Meneses do not explain precisely how a constitutional violation is also a statutory violation. However, it appears they rely on CalECPA‘s requirement that a warrant must comply with all “provisions of California and federal law” (
The judgments are affirmed.
PERLUSS, P. J.
We concur:
SEGAL, J.
FEUER, J.