AFFIRMED and Opinion Filed August 23, 2023
S In The
Court of Appeals
Fifth District of Texas at Dallas
No. 05-21-00855-CR
AARON RAYSHAN WELLS, Appellant
V.
THE STATE OF TEXAS, Appellee
On Appeal from the Criminal District Court No. 4
Dallas County, Texas
Trial Court Cause No. F19-75986
OPINION
Before Justices Reichek and Goldstein1
Opinion by Justice Reichek
Aaron Rayshan Wells appeals his conviction for capital murder. Bringing six
issues, appellant contends (1) he was not brought to trial timely under the Interstate
Agreed Detainer Act (IADA), (2) the trial court abused its discretion in denying his
motion to suppress evidence obtained through a geofence warrant, (3) the State
failed to show the Google data used as a basis for its expert’s testimony was reliable,
(4) the trial court abused its discretion in denying his motion for continuance, (5) the
1
Justice David Schenck was a member of the original panel, but Justice Schenck is no longer a member
of the Court, and he did not participate in the issuance of this opinion.
evidence was legally insufficient to support his conviction, and (6) the submission
of a jury instruction on conspirator liability constituted harmful error. We affirm the
trial court’s judgment.
Background
We begin with a brief overview of the facts concerning the offense and
investigation. In the early morning hours of June 24, 2018, Nikita Dickerson
engaged in her nightly routine of meeting her boyfriend, Jimmy Giddings, in the
driveway of their house to escort him inside. Giddings was a drug dealer, and
Dickerson brought a gun with her for their protection. Dickerson stated she was not
aware of any particular threats to Giddings, but the neighborhood was unsafe.
On this occasion, as Giddings was getting out of his car, a group of men rushed
toward him from the vicinity of a church across the street. One of the men shot at
Dickerson multiple times, and she sustained non-fatal injuries. As Dickerson
collapsed to the ground, she dropped her gun, which was picked up by the man who
shot her. Security camera recordings show Giddings running into the house and
closing a metal gate in the entryway behind him. The men followed, kicking open
the gate to gain entry. Three men are clearly visible in the recording with the lower
parts of their faces covered. The men entered the house, forcing Dickerson to get up
and walk inside with them at gunpoint. A short time later, a fourth man, later
identified as appellant, ran past the camera into the house.
–2–
Once in the house, the men demanded money. During the course of the
robbery, Giddings was shot and killed. One of Dickerson’s children, who was in a
bedroom of the house during the offense, called 911. The men fled before police
arrived.
A homicide detective, Jeffrey Loeb, released still pictures of the three men
who could be seen clearly in the security camera footage. A public tip line was
opened to try to identify them. Although the lower portions of the men’s faces were
covered, the pictures showed distinctive tattoos.
When no productive leads were generated by the tip line, Detective Loeb
requested a geofence search warrant to obtain information from Google about
cellular devices located in the area at the time of the offense. Based on the security
camera recording timestamp and footage showing that the men were in the area of
the church immediately before and after the offense, Loeb obtained a warrant to
search Google’s records for information on devices located within a rectangular
geofence encompassing Giddings’s house and the portion of the church directly
across the street between 2:45 a.m. and 3:10 a.m. on June 24. Ultimately, a cellular
phone associated with appellant was identified as being at the scene. Through
appellant’s phone records and a search of social media, police were able to identify
Milton Prentice, Brian Groom, and Kiante Watkins as the other three men involved
in the offense.
–3–
Appellant was found incarcerated for a different offense in a federal prison in
Tennessee. After being transferred to Dallas County, he was tried for and convicted
of capital murder. This appeal followed.
Analysis
I. IADA
In his first issue, appellant contends the trial court erred in denying his motion
to dismiss under the IADA because the State of Texas failed to bring him to trial
within 180 days after he requested disposition of his case. The IADA is a
congressionally sanctioned compact between the United States and the states that
have adopted it, including Texas. See Kirvin v. State, 394 S.W.3d 550, 555 (Tex.
App.—Dallas 2011, no pet.). The Act outlines a cooperative procedure to be used
when a state is seeking to try a prisoner who is being held in a penal or correctional
institution of another state. See TEX. CODE CRIM. PROC. ANN. art. 51.14. The state
with the untried indictment, information, or complaint must file a detainer with the
institution that is holding the prisoner. State v. Votta, 299 S.W.3d 130, 135 (Tex.
Crim. App. 2009).2 The institution must then promptly inform the prisoner the
detainer has been filed and that he has the right to request final disposition of the
charges made the subject of the detainer. Id. The prisoner may make such a request
2
A detainer is a request that is filed by a criminal justice agency with the institution in which the
prisoner is incarcerated asking that the prisoner be held for the agency, or that the agency be advised when
the prisoner’s release is imminent. Id. at 135, n. 5.
–4–
by giving written notice to the warden of the facility in which he is being held who
is then required to forward the request to the appropriate court and prosecuting
officer. Id. Once the request for final disposition has been received by the court and
prosecuting officer, the prisoner must be brought to trial within 180 days unless a
continuance is granted under the Act. Id.
A continuance granted under the Act must be (1) by a court of competent
jurisdiction, (2) in open court, (3) with the defendant or his counsel present, (4) for
good cause, and (5) be necessary or reasonable. Kirvin, 394 S.W.3d at 556. The
“open court” requirement does not mandate a formal proceeding, but is intended to
prevent ex parte or sua sponte continuances. Id. Where a continuance is agreed to
by both sides, it is deemed to be reasonable, necessary, and granted for good cause.
Id. at 556–57. The defendant’s personal consent to the continuance is not required;
his counsel’s signature is sufficient. Id. at 556. Agreed continuances toll the
statutory period within which the defendant must be tried. Id.
The record in this case shows the State of Texas filed its detainer with the
Tennessee prison and, on April 8, 2020, appellant signed his request for final
disposition. Appellant’s request was received by the trial court on May 8. The State
asserts the district attorney did not receive the disposition request until May 29.
Appellant was transferred to a Dallas County jail on June 30.
On July 15 and August 7, both the district attorney and appellant’s defense
counsel signed agreed requests for continuance, otherwise known as “pass slips,”
–5–
resulting in a trial setting in September. On September 23, the parties filed a joint
motion for continuance stating that, although the IADA required appellant to be tried
within 180 days, the postponement of jury trials due to the pandemic constituted
good cause to continue the trial date in this case. Based on the joint motion, the trial
court reset the trial date to March 1, 2021.
Between February 11, 2021 and April 15, 2021, the district attorney and
defense counsel jointly requested five more continuances resulting in a trial setting
of August 19. On June 25, the court conducted a pre-trial conference at which the
State, appellant, and his counsel appeared. The State announced at the hearing that
it was ready to proceed to trial. Defense counsel, however, stated she had a conflict
and requested the trial be moved to the next available date on the court’s jury
calendar, which was September 21.
Appellant spoke on his own behalf and stated he had two concerns. His first
concern was that he felt there had been a lack of communication between him and
his counsel, resulting in her failure to file motions or “come up with a defense”
despite the fact the case was set for trial. Because of this, appellant requested he be
appointed new counsel. Appellant’s second concern was the delay in his case going
to trial, and he requested the charges against him be dismissed under the IADA.
The trial court denied appellant’s request for new counsel and appellant does
not challenge the denial on appeal or assert he received ineffective assistance. In
response to appellant’s IADA complaint, the court agreed appellant “certainly
–6–
need[ed] to have this case tried.” The court went on to find good cause existed to
extend the trial date to September 21, but cautioned that it would be the final
continuance granted. Jury selection began on September 21, and trial commenced
on September 23.
As stated above, agreed continuances toll the statutory period within which a
defendant must be brought to trial under the IADA. Id. at 557. The record here
shows the State and defense agreed to multiple continuances of the trial date
covering a period of 379 days between July 15, 2020 and August 19, 2021. During
a large portion of this time, jury trials were not being conducted due to the global
pandemic. Defense counsel then requested an additional continuance which tolled
the statutory period for a further 35 days. See id. at 556 (reasonable delays caused
by defense motions toll statutory period). Appellant’s trial commenced 482 days
after his request for disposition was received by the district attorney and 503 days
after his request was received by the trial court. Even assuming the latter date
controls, 414 of the 503 days were tolled, meaning that only 89 days of the 180-day
period ran from the date of the request to the beginning of trial. Appellant’s trial
was, therefore, timely under the IADA. Id. at 558. We resolve appellant’s first issue
against him.
II. Geofence Search
In his second issue, appellant contends the trial court abused its discretion in
refusing to suppress location history data obtained from Google with a geofence
–7–
search warrant. A geofence warrant allows police investigators to search location
history data for compatible mobile devices located within a specified area during a
specified period of time. In re Search Warrant Application for Geofence Location
Data Stored at Google Concerning an Arson Investigation, 497 F. Supp. 3d 345, 351
(N.D. Ill. 2020) (“Arson”). This type of warrant is essentially the reverse of a global
positioning systems (“GPS”) warrant which allows a search of location data
generated by a specific device belonging to a person known or suspected to have
been involved in criminal activity. See In re Search of Info. that is Stored at
Premises Controlled by Google LLC, 579 F. Supp. 3d 62, 69 (D.D.C. 2021) (“D.C.
Google I”). With a geofence warrant, police investigators identify the geographic
area in which criminal activity occurred and seek to identify device users at that
location when the crime was committed. Arson, 497 F. Supp. 3d at 351. Google
calculates the location of a device that has enabled Google location history using
input from cell towers, GPS, and signals from nearby wireless internet networks
(“Wi-Fi”) and Bluetooth beacons. United States v. Rhine, No. 21-0687, 2023 WL
372044, at *17 (D.D.C. Jan. 24, 2023). Because Google location history includes
multiple inputs, it is “considerably more precise than other kinds of location data.”
Id. For each device, Google retains subscriber information which may include the
subscriber’s name, address, telephone number, and other identifiers. Arson, 497 F.
Supp. 3d at 351. Law enforcement officers use a geofence search warrant to seize
–8–
this data using a multi-step process to identify criminal suspects and potential
witnesses to the crime. See D.C. Google I, 579 F. Supp. 3rd at 71–72.
In this case, Detective Loeb submitted a warrant application outlining a three-
step search process. In the first step, Google would be asked to create an anonymized
list of all devices located within the “target location” during the time period of 2:45
a.m. to 3:10 a.m. on June 24, 2018. The application defined the target location using
four latitude and longitude coordinates connected by straight lines and included a
visual reference image of the search area.
As shown, the search area was limited to the house where the offense occurred and
a portion of the church property across the street.
Once the anonymized list was produced, law enforcement would then analyze
the data to identify users who may have witnessed or participated in the offense. For
–9–
those users identified as relevant to the investigation, Google would then provide
additional anonymized location history outside of the target location for a period not
to exceed sixty minutes before and after the last timestamp associated with the device
within the target location. The purpose of this additional information was to
eliminate users who, based on the contextual data, did not appear to fall within the
scope of the warrant. For the accounts determined to be relevant after this narrowing
process, Google would then provide the subscriber information, including the user’s
name and email address.
In addition to the search description, the warrant application included
background information on Google’s location services, discussion of the prevalence
of Google accounts on cellular phones, and a probable cause statement by Loeb. The
probable cause statement laid out the basic facts of the offense. It was noted that
surveillance video obtained from the church showed the suspects gathering in the
parking lot behind one of the church buildings across the street from the victim’s
house just prior to the offense and showed the suspects fleeing immediately
afterwards in a car that was left parked in the lot. Loeb stated it was likely that at
least one of the four suspects had a device on him during the commission of the
offense that had enabled Google location services. He opined that “[i]t is common
practice that home invasion robbery suspects keep an open line with someone
outside of the residence while committing this type of offense to keep an eye out for
responding police officers.” It was also noted that the initial shooting of Dickerson
–10–
occurred outside the house in the front yard, and the police were looking not only
for persons involved in the offense, but also for those who were present at the time
as potential witnesses.
A district judge approved the warrant and, as a result of the initial search,
Google identified three devices in the target location at the time of the offense. Once
the search was expanded, Loeb narrowed the search to a single Google account that
“was determined to belong to a suspect who clearly had involvement in this offense.”
The account belonged to appellant and, with appellant’s name, email address, and
Google account identification number, Loeb obtained a second warrant for
appellant’s Google account information. Appellant does not challenge this second
warrant other than as the “fruit of the poisonous tree.”
Google account records provided in response to the second warrant showed
that appellant had changed his phone number since he created his Google account.
A woman using appellant’s former phone number gave Loeb appellant’s new
number. A warrant was then issued to T-Mobile for appellant’s phone records,
including cell-site location information. The phone records confirmed that appellant
was in the area at the time of the offense. The records also showed the persons with
whom appellant was in communication around the time of the offense. Through
social media, detectives were able to identify Prentice, Groom, and Watkins as the
other three men involved in the offense based on their distinctive tattoos.
–11–
Prior to trial, appellant moved to suppress the information obtained through
the geofence warrant. Appellant argued the search was an unreasonably broad
invasion of privacy and unsupported by probable cause. The State responded that
appellant did not have a reasonable expectation of privacy in the State’s limited
search of his phone’s location history. The State additionally contended the motion
should be denied because the warrant was valid and, alternatively, suppression of
the evidence was inappropriate under the good faith exception to the exclusionary
rule. The trial court denied appellant’s motion to suppress, and the evidence was
admitted.
On appeal, appellant argues the State’s search pursuant to the geofence
warrant was a violation of the Fourth Amendment’s constitutional guarantee against
unreasonable searches and seizures. See U.S. CONST. amend. IV. The ultimate
measure of the constitutionality of a governmental search is its “reasonableness.”
Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 652 (1995). Whether a particular
search is reasonable is judged by balancing its intrusion on the individual’s Fourth
Amendment interests against its promotion of legitimate governmental interests. Id.
at 652–53. Where a search is conducted by law enforcement officials to discover
evidence of criminal wrongdoing, reasonableness generally requires a warrant. Id.
at 653. The United States Supreme Court has determined that, to be compliant with
the Fourth Amendment, a warrant must meet three requirements: (1) it must be
issued by a neutral, disinterested magistrate; (2) it must be supported by probable
–12–
cause that the evidence sought will aid in a “particular apprehension or conviction
for a particular offense”; and (3) it must particularly describe the things to be seized
and the place to be searched. Dalia v. United States, 441 U.S. 238, 255 (1979).
Appellant asserts the State’s geofence warrant in this case failed to meet the second
and third requirements.
The issuance of a warrant is supported by probable cause if, under the totality
of the circumstances set forth in the affidavit before the magistrate, there is a fair
probability that contraband or evidence of a crime will be found in a particular place
at the time the warrant is issued. State v. Jordan, 342 S.W.3d 565, 568–69 (Tex.
Crim. App. 2011). “The magistrate may interpret the affidavit in a non-technical,
common-sense manner, and may draw reasonable inferences from the facts and
circumstances contained within its four corners.” Id. at 569. We give great
deference to the magistrate’s determination of probable cause and, on review, only
ensure the magistrate had a substantial basis for concluding probable cause existed.
Id.
The particularity requirement is related to the probable cause requirement
because it allows the magistrate to determine whether probable cause exists for the
requested search. Bonds v. State, 403 S.W.3d 867, 875 (Tex. Crim. App. 2013).
Among the objectives of requiring particularity with respect to the place to be
searched and things to be seized are: (1) limiting the officer’s discretion by
narrowing the scope of the search, and (2) minimizing the danger of searching the
–13–
person or property of an innocent bystander or property owner. Id. at 874–75.
Primarily, the requirement is “meant to prevent general searches and the seizure of
one thing under a warrant that describes another thing to be seized.” State v. Powell,
306 S.W.3d 761, 765 (Tex. Crim. App. 2010).
Geofence warrants are a relatively new addition to the array of investigative
tools available to law enforcement. Because of this, the amount of case law on this
type of warrant is limited. See Rhine, 2023 WL 372044, at *17. But, while the cases
are not numerous, those that have addressed the constitutional issues raised by
geofence warrants have done so in depth. See e.g. id. at *21–27 (surveying geofence
warrant cases). We find the reasoning in these opinions to be consistent, persuasive,
and instructive on the warrant presented here. See Price v. Super. Ct. of Riverside
Cnty, No. E078954, 2023 WL 4312776, at *11 (Cal. App. 4th July 3, 2023).
The case of Price v. Superior Court of Riverside County has strikingly similar
facts to those presented in this case. In Price, a man was shot on the front porch of
his home in a residential area. Id. at *3. The man’s brother was a witness to the
offense and saw two men flee the scene in opposite directions — one ran north, while
the other got into a vehicle and headed south. Id. Surveillance video from a gas
station nearby showed the same vehicle later heading east on a different street. Id.
at *4.
A detective obtained a warrant seeking location history information from
Google for devices located within a geofence encompassing the front yard of the
–14–
victim’s house, including the front porch area where the shooting occurred, and the
street in front of the house for the lengths of two houses heading north and south.
Id. at *5. Based on the timing of multiple 911 calls received about the offense, the
time period specified for the first stage of the search was from 10:00 p.m. to 10:22
p.m. Id. In his affidavit in support of the warrant, the detective averred that “in his
experience people who plan and commit crimes together use cell phones to
communicate.” Id. at *11.
The first search revealed five devices within the geofence during the twenty-
two-minute time period. Id. at *5. The second stage of the search provided a “more
expanded view” of the location and movements of the five devices and indicated that
two of the devices travelled past the gas station where the suspects’ car was recorded
on video surveillance. Id. at *4–*5. The third stage of the search revealed the
identity of the Google account subscribers for the two devices identified as relevant.
Id. at *5. Additional warrants were then secured to obtain the Google account and
cellular phone records for the suspects. Id.
In its review of the constitutionality of the geofence warrant, the court in Price
concluded the warrant was both supported by probable cause and a “model of
particularity.” Id. at *14. Given the specificity of the geofence and the “indisputable
common knowledge that most people carry cell phones virtually all the time,” the
court reasoned there was at least a “fair probability” that the suspects in the murder
would be identified in the search. Id. at *12. The court went on to state that
–15–
if a geofence warrant is narrowly tailored, in its initial search
parameters, or geographic scope and time period, to maximize the
probability it will capture only suspects and witnesses, and to minimize
searches of location data and identifying information of individuals for
whom there is no probable cause to believe [they] were suspects or
witnesses (uninvolved individuals), then the discretion afforded to the
executing officer by Google’s multi-step production protocol will be
constitutionally immaterial.
Id. at *13.
The geofence warrant in this case was equally likely to reveal evidence of the
identities of suspects and potential witnesses to the offense, and it was even more
likely than the warrant in Price to exclude uninvolved persons. Like the warrant in
Price, the geofence here encompassed only the area where the suspects were known
to have been at the time of the offense. Also like Price, the area was residential in
nature, made up largely of single family homes, and was not heavily trafficked at the
time. Because the period for the initial search in this case was between 2:45 a.m.
and 3:10 a.m., there was a significantly diminished possibility that anyone
uninvolved would be located in or around the search area, which included only the
victim’s house and a then-closed church building. Additionally, because one of the
shootings took place outside the house, there was a strong possibility that everyone
within the geofence would have witnessed at least part of the offense.
Although it is unclear whether or to what extent the second stage of the
location history search in Price was limited in time, the warrant here limited the
anonymized information produced in the second stage to only location history for a
period of sixty minutes before and after the initial search period, for a maximum
–16–
time of approximately two and one-half hours. This limited time-frame in the early
morning hours of a single day would not provide nearly the “all-encompassing
record” or “intimate window into a person’s life” that concerned the United States
Supreme Court when it addressed the seizure of one hundred and twenty-seven days
of historical cell-site information in Carpenter v. United States, 138 S. Ct. 2206,
2217 (2018).
Appellant acknowledges that the three-step narrowing process used in the
State’s geofence warrant “effectively filtered out the innocent,” but argues the initial
search impermissibly included innocent third parties merely because they were
present in an area where a crime was committed. As we have already established,
however, the warrant in this case, even in its initial stage, was unlikely to identify
persons who were not either a suspect or witness to the offense. Furthermore, a
search is not constitutionally impermissible simply because it may infringe on the
privacy interests of uninvolved third persons. Rhine, 2023 WL 372044, at *26; D.C.,
579 F. Supp. 3d at 84; Arson, 497 F. Supp. 3d at 361. It is sufficient that the warrant
did not have the potential for revealing the personal information of a substantial
number of uninvolved persons. Rhine, 2023 WL 372044, at *26.
Appellant argues that, to demonstrate probable cause for the warrant, the State
was required to provide specific evidence showing the suspects were carrying cell
phones with enabled Google location services and that the phones would contain
evidence of the crime. As multiple courts, including the United States Supreme
–17–
Court, have recognized—cell phones are ubiquitous. See Carpenter, 138 S. Ct. at
2218. Judges are not required to “check their common sense at the door and ignore
the fact that most people compulsively carry cell phones with them all the time.”
United States v. James, 3 F. 4th 1102, 1105 (8th Cir. 2021). Detective Loeb’s
affidavit explained that, in his training and experience, co-conspirators in a home-
invasion robbery use phones to keep an open line with someone outside the
residence. The affidavit further stated that “nearly every cellular phone using the
Android operating system has an associated Google account” and non-Android
devices may also have Google accounts that are used by the company for location-
based services, advertising, and search results. Although it is possible the suspects
were not carrying cell phones with enabled Google location services during the
offense, probable cause is about “fair probabilities,” not near certainties. Id.; see
also Arson, 497 F. Supp. 3d at 355 (“probable cause does not require conclusive
evidence that links a particular place or item to a crime”). As for establishing how
the cell phone would contain evidence of the offense, the affidavit in support of the
warrant made clear that police were seeking the identities of persons who were at
the scene through use of the phone’s location data. See Price, 2023 WL 4312776,
at *11.
Appellant relies heavily on the case of United States v. Chatrie, 590
F.Supp.3rd 901 (E.D. Va. 2022), to argue the State’s geofence warrant was
overbroad. Appellant’s reliance is misplaced. In Chatrie, the court held a geofence
–18–
warrant was impermissibly overbroad because it failed to establish particularized
probable cause to search every Google user within the geofence. Id. at 929. The
geofence at issue encompassed 17.5 acres in a commercial area between the hours
of 4:20 p.m and 5:20 p.m. Id. at 922. The court concluded the warrant was
“completely devoid of any suggestion that all, or even a substantial number of the
individuals searched had participated in or witnessed the crime.” Id. at 929. In
contrast, the geofence warrant before us was as narrowly tailored as possible to
capture only location data for suspects and potential witnesses.
The geofence warrant cases to date can generally be divided into two
categories— those in which the geofence search warrant was found constitutionally
infirm because it was not sufficiently limited as to time and place so as to restrict the
executing officer’s discretion and minimize the danger of searching uninvolved
persons,3 and those in which the warrant satisfied the Fourth Amendment because it
established probable cause to search every person found within the geofence area.4
Based on the analysis above, we conclude the warrant here falls into the second
3
People v. Meza, No. B318310, 2023 WL 5287224, at *11 (Cal. Ct. App. 2023); Chatrie, 590 F. Supp.
3d at 929; In re Search of Info. that is Stored at the Premises Controlled by Google (“Kansas”), 542 F.
Supp. 3d 1153, 1158 (D. Kan. 2021); In re Search of Info. Stored at Premises Controlled by Google
(”Pharma I”), No. 20 M 297, 2020 WL 5491763, at *7 (N.D. Ill. July 8, 2020); In re Search of Info. Stored
at Premises Controlled by Google (“Pharma II”), 481 F. Supp. 3d 730, 756 (N.D. Ill. 2020).
4
Price, 2023 WL 4312776, at *18; In re Search of Info. that is Stored at Premises Controlled by Google
(“Texas”), No. 2:22-mj-01325, 2023 WL 2236493, at *14 (S.D. Tex. Feb. 14, 2023); United States v. Smith,
No. 3:21-cr-107-SA, 2023 WL 1930747, at *10 (N.D. Miss. Feb. 10, 2023) (warrant not overbroad, but
officers failed to follow process set out in warrant); Rhine, 2023 WL 372044, at *32 (warrant not overbroad,
but second warrant used to obtain de-anonymized information); D.C. Google I, 579 F. Supp. 3d at 80–81;
Arson, 497 F. Supp. 3d at 363.
–19–
category. Even when the geofence warrant was found to be invalid, however, courts
have refused to exclude evidence resulting from the search. This is because evidence
obtained in objective good faith reliance upon a warrant that is valid on its face is an
exception to the exclusionary rule. See United States v. Leon, 468 U.S. 897, 922
(1984); see also TEX. CODE CRIM. PROC. ANN. art. 38.23(b); McClintock v. State,
541 S.W.3d 63, 73 (Tex. Crim. App. 2017). “Given the dearth of authority directly
on point and the novelty of the particular surveillance technique at issue,” courts
have determined evidence obtained through an invalid geofence warrant was
nevertheless admissible because it was not “objectively unreasonable” for officers
to believe the warrant was valid. Meza, 307 Cal. Rptr. 3d at 544; Chatrie, 590 F.
Supp. 3d at 938. Appellant here makes no argument that it was objectively
unreasonable for Detective Loeb to rely on the geofence warrant to obtain his
location history. Accordingly, even if the warrant was invalid, appellant has failed
to show the trial court erred in refusing to grant his motion to suppress.
Because we conclude the warrant at issue satisfies the requirements of the
Fourth Amendment and, alternatively, Detective Loeb’s reliance on the warrant was
objectively reasonable, it is unnecessary for us to address the State’s argument that
appellant had no reasonable expectation of privacy in his location history.5 We
resolve appellant’s second issue against him.
5
We note that no case has been willing to go as far as the State suggests and hold that law enforcement
officers do not need to obtain a warrant before searching Google’s location history data stores. The State
–20–
III. Google Location Data Reliability
In his third issue, appellant contends the trial court abused its discretion in
allowing the State’s mapping expert, Greg Gambrell, to testify about appellant’s
location history based, in part, on the data obtained from Google. Appellant argues
the State failed to show the Google data was reliable. Trial courts are given wide
latitude in their rulings on the reliability of expert testimony. Wilson v. Shanti, 333
S.W.3d 909, 913 (Tex. App.—Houston [1st Dist.] 2011, pet. denied). If there is
evidence supporting the trial court’s decision to admit the challenged evidence, there
is no abuse of discretion and we must defer to that decision. Osbourn v. State, 92
S.W.3d 531, 538 (Tex. Crim. App. 2002).
In a hearing outside the presence of the jury, Gambrell testified he was a
graphic digital multimedia evidence specialist who was certified in call detail
records and geolocation analysis. To receive the certification, Gambrell completed
five years of training through the FBI, the United States Secret Service, and the
National White Collar Crime Center. Gambrell also completed approximately one
cites Sims v. State in which the Texas Court of Criminal Appeals held a defendant did not have a legitimate
expectation of privacy in his physical movements or location as revealed by three hours of real-time
surveillance done by “pinging” his phone. See Sims v. State, 569 S.W.3d 634, 645 (Tex. Crim. App. 2019).
As the U.S. Supreme Court has observed, however, historical location information maintained by a third
party “does not fit neatly under existing precedents.” Carpenter, 138 S. Ct. at 2214. Law enforcement’s
real-time tracking of a known, and potentially dangerous, suspect, and its search of a vault of historical
location data to determine the whereabouts of unknown persons and innocent bystanders, are fundamentally
different things. We share the concern of other courts that “current Fourth Amendment doctrine may be
materially lagging behind technological innovations.” Chatrie, 590 F. Supp. 3d at 925. And we bear in
mind that the central aim of the Framers was “to place obstacles in the way of a too permeating police
surveillance.” Carpenter, 138 S. Ct. at 2214 (quoting United States v. Di Re, 332 U.S. 581, 595 (1948)).
–21–
hundred and twenty hours of training on the use of ZetX software, which creates
illustrations of location histories.
Gambrell testified he input thousands of lines of appellant’s location history
data, obtained from both Google and appellant’s phone service provider, into the
ZetX program to create maps showing appellant’s movements. Gambrell stated the
records he received from Google were certified and ZetX provides a failsafe that
detects if records have been altered. Google additionally provided a business records
affidavit stating that Google servers recorded the data automatically at the time, or
reasonably soon after, it was transmitted by appellant’s device, and the data was kept
in the ordinary course of business. The Google affidavit averred the records it
produced were “a true duplicate of original records that were generated by Google’s
electronic process or system that produces an accurate result” and “[t]he accuracy of
Google’s electronic process and system is regularly verified by Google.”
Gambrell discussed how the Google data was collected using Wi-Fi, GPS, and
cell towers, and he stated the spreadsheets of location data provided by Google
specify which of these sources produced each data point along with its range of
accuracy. According to Gambrell, the range of accuracy is within ten meters for
GPS, within one hundred meters for Wi-Fi, and within miles for cell towers.
Gambrell elaborated on how each of these technologies collects location data from
cellular devices and testified that all three are considered valid sciences accepted
among the scientific community. He further stated Google relies on the accurate
–22–
collection of location information as part of its business model to provide location
services and targeted advertising.
With respect to the specific Google data used in this case, Gambrell testified
he verified its accuracy in several ways. First, he compared the location data
received from Google to the cell-site location data received from appellant’s phone
service provider to determine if there were any conflicting data points. Next, he
looked to other evidence in the case, such as the surveillance videos, to confirm that
it correlated with the Google data. He also looked within the Google data itself for
anomalies or outliers that would indicate it was inaccurate. As a final confirmation,
Gambrell manually validated the accuracy range of the Google data points from the
crime scene using a measuring wheel. At the conclusion of the hearing, the trial
court ruled that Gambrell’s testimony, and the State’s demonstrative evidence
regarding appellant’s movements, were admissible.
Appellant contends that Gambrell’s testimony did not meet the standard of
reliability for expert testimony set out in Kelly v. State, 824 S.W.2d 568, 573 (Tex.
1992).6 For expert testimony to be reliable under Kelly, (1) the underlying scientific
theory must be valid, (2) the technique applying the theory must be valid, and (3)
the technique must have been properly applied on the occasion in question. Id.
6
The State responds that Gambrell’s testimony was not based on hard science, but rather on specialized
technical training and, therefore, the more flexible standard of reliability set forth in Nenno v. State applies.
See Nenno v. State, 970 S.W.2d 549, 561 (Tex. Crim. App. 1998), overruled on other grounds by State v.
Terrazas, 4 S.W.3d 720 (Tex. Crim. App. 1999). Because we conclude Gambrell’s testimony met the
standards set forth in Kelly, we need not address this argument.
–23–
Whether a particular field of study is legitimate is the type of question that can be
resolved as a general matter so that courts can take judicial notice of the reliability
of the type of evidence at issue. Morris v. State, 361 S.W.3d 649, 655 (Tex. Crim.
App. 2011). Trial courts are not required to “reinvent the scientific wheel in every
trial.” Id.
The evidence the State sought to introduce through Gambrell was maps of
appellant’s movements in the days leading up to, including, and immediately after
the offense. The maps were created using cell-site location data obtained from
appellant’s phone service provider and GPS, Wi-Fi, and cell-site location data
obtained from Google. Appellant challenges only the reliability of the Google
location data.
As the State informed the trial court at the hearing, this Court and many others
have already concluded that maps based solely on cell-site location data, including
specifically ZetX mapping, are sufficiently reliable to be admissible at trial. See
Trevino v. State, No. 05-19-00295-CR, 2020 WL 2537246, at *8 (Tex. App.—Dallas
May 19, 2020, no pet.) (mem op., not designated for publication); Patrick v. State,
No. 05-18-00435-CR, 2018 WL 3968781, at *30 (Tex. App.—Dallas Aug. 20, 2018
no pet.) (mem op., not designated for publication) (citing multiple other cases);
Thompson v. State, 425 S.W.3d 480, 489 (Tex. App.—Houston [1st Dist.] 2012, pet.
ref’d); see also Carpenter, 138 S.Ct at 2216 (cell-site location information provides
a detailed and comprehensive record of a person’s movements); United States v.
–24–
Schaffer, 439 F. App’x. 344, 347 (5th. Cir. 2011) (field of historical cell-site analysis
neither untested nor unestablished). Given that a map based solely on cell-site
location information from appellant’s phone service provider would have been
admissible, appellant fails to explain how a map that combines this same data with
the location data provided by Google is rendered less reliable. Indeed, the
combination of technologies used by both companies makes it significantly more
reliable. See State v. Pierce, 222 A.3d 582, 590 (Del. Super. Ct. 2019) (accuracy of
Google Wi-Fi location data verified by other mechanisms such as GPS).
In addition to gathering location information through cell towers, Google also
mines data from GPS and Wi-Fi routers. GPS in particular is significantly more
precise than cell-site location information and has been found by this Court to be
reliable. See Brown v. State, 163 S.W.3d 818, 824 (Tex. App.—Dallas 2005, pet.
ref’d); see also United States v. Beverly, 943 F.3d 225, 230 n.2 (5th Cir. 2019).
Appellant acknowledges this fact in his argument in support of his second issue.
In State v. Pierce, the Superior Court of Delaware thoroughly analyzed the
reliability of Google’s Wi-Fi location data under the standard set out in Daubert v.
Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). The Daubert standard is
“virtually identical” to the Kelly standard. See Hartman v. State, 946 S.W.2d 60, 62
(Tex. Crim. App. 1997) (en banc). The court in Pierce concluded that Google’s
location data met all the criteria for reliability, including wide acceptance in the
scientific community. Pierce, 222 A.3d. at 591. The fact that Google uses the
–25–
location data for commercial purposes is a recognized indicator of reliability. Id.;
see also Carpenter, 138 S. Ct. at 2217.
In addition to the general reliability of Google’s location data, Gambrell
independently verified the accuracy of the information he was provided in this case.
His verification of the Google location data included corroborating it using
appellant’s cell phone records and other evidence in the case, as well as manually
assessing the accuracy of certain data points using a measuring wheel. This type of
independent corroboration supports the trial court’s finding of reliability. See Alyea
v. State, No. 14-19-00498-CR, 2021 WL 5117972, at *3–5 (Tex. App.—Houston
[14th Dist.] Nov. 4, 2021, pet. ref’d) (mem. op., not designated for publication).
Appellant argues the State offered no evidence to show “Google’s process for
culling and delivering data in response to a warrant or court order.” As observed by
the court in Chatrie, Google is always collecting location data and stores all of the
data it collects in a vast “Sensorvault.” Chatrie, 590 F.Supp.3d at 909. The warrant
served on Google in this case ordered the company to turn over “all location data”
associated with appellant’s Google account for the time period of June 1, 2018 to
June 30, 2018, including GPS, Wi-Fi, or Bluetooth sourced location history data
generated from devices associated with appellant’s Google account, and all data
“collected or maintained by Google Location Services and/or Google Location
History and any stored location data from or derived from GPS, Wi-Fi access points,
cell tower triangulation/trilateration or other measurement” related to appellant’s
–26–
Google account. There is nothing in the record to suggest Google did not turn over
all the information the State requested.
Appellant points to the testimony of his defense expert, Steve Watson,
concerning the unreliability of Google’s data. Most, if not all, of this testimony was
offered by the defense after the trial court had already ruled on the admissibility of
Gambrell’s testimony. Furthermore, Watson acknowledged the accuracy of the
independent technologies relied upon by Google—GPS, Wi-Fi, and cell tower
data—to determine location. Watson’s focus was on the proprietary, and therefore
un-reviewed, algorithm used by Google to combine those inputs to produce the
location determination it uses in its services. The data provided by Google in
response to the warrant did not employ the algorithm, however. It provided only the
raw data points from each of the different sources of location information separately
along with their ranges of accuracy. The ranges of accuracy for each data point were
fully taken into account by Gambrell and explained to the jury.
We conclude the case law concerning the technology at issue, and the
evidence concerning its specific application and reliability in this case, were
sufficient to support the trial court’s decision to admit Gambrell’s testimony.
Accordingly, the trial court did not abuse its discretion and we overrule appellant’s
third issue.
–27–
IV. Motion for Continuance
In his fourth issue, appellant contends “the trial court abused its discretion
when it denied his motion for continuance based, in part, on evidence the State did
not disclose under Brady v. Maryland, 373 U.S. 83 (1963).” He asserts the State
failed to turn over evidence relating to the “culpability” of Dickerson, Giddings’s
girlfriend, and a man whose fingerprint was found on one of the cars at the crime
scene. Appellant argues he “undeniably suffered harm because he could not explore
the viable theory that Dickerson masterminded [Giddings’s] murder.” We review a
trial court’s ruling on a motion for continuance for an abuse of discretion. Cruz v.
State, 565 S.W.3d 379, 381 (Tex. App.—San Antonio 2018, no pet.).
At the hearing on appellant’s motion for a continuance, Jimmy Spurger, an
investigator with the public defender’s office, testified he had uncovered information
about Dickerson after looking into tips received by the Dallas Police Department.
Spurger testified these tips were provided to the defense in July 2020, but he did not
begin investigating them until September 15, 2021, a few days before the hearing.
The first tip stated that Dickerson had Giddings murder her previous husband
“Nathaniel” so she could benefit financially. This person also believed Dickerson
arranged to have Giddings murdered. On the same day, police received a second tip
stating that Dickerson was “in on the robbery” and “she was shot to make it look
good.”
–28–
Spurger began researching Dickerson and found a relative named Nathan
Williams who was murdered in 2012. A CLEAR report on Williams stated he was
associated with both Dickerson and Giddings and showed him living at the house
where Giddings was killed from September 2016 to January 2018, well after his
death in 2012. A man named Ghiri Johnson was also listed as an “associate” of
Williams and a “relative/associate” of Dickerson and Giddings. A fingerprint of
Johnson’s was found at the scene of Giddings’s murder. Spurger requested a copy
of the file on William’s murder, and the State turned it over the same day.
At the hearing on appellant’s motion, defense counsel stated she wanted to
make it clear she was not accusing the State of hiding anything. Her argument was
only that the State had failed in its obligation to investigate these tips and they
“should have found the Brady.” Counsel asked for additional time to review the file
on Williams’s murder. The trial court denied the motion.
A Brady violation occurs when the State suppresses, willfully or
inadvertently, evidence favorable to the appellant. Harm v. State, 183 S.W.3d 403,
406 (Tex. Crim. App. 2006) (en banc). Brady does not require the State to disclose
exculpatory information that it does not know exists. Id. at 407. Nor does the State
have any duty to seek out such information independently on the defendant’s behalf.
Duncan v. State, No. 05-96-01027-CR, 1997 WL 691438, at *6 (Tex. App.—Dallas
Nov. 6, 1997, pet. ref’d) (not designated for publication); Palmer v. State, 902
S.W.2d 561, 563 (Tex. App.—Houston [1st Dist.] 1995, no pet). If the defendant,
–29–
using reasonable diligence, could have obtained the information about which he
complains, there is no Brady violation. Duncan, 1997 WL 691438, at *6; Westley
v. Johnson, 83 F.3d 714, 726 (5th Cir. 1996).
Here, the record shows the State did not withhold or suppress evidence. The
State simply did not pursue two tips it received. Those tips were timely provided to
the defense. Although the defense was in possession of the tips for more than a year,
it did not begin to investigate them until shortly before trial. We conclude appellant
has failed to show a Brady violation and the trial court did not abuse its discretion in
denying appellant’s motion for a continuance. We resolve appellant’s fourth issue
against him.
V. Accomplice Witness Testimony
Appellant’s fifth issue challenges the sufficiency of the evidence to support
his conviction. Appellant argues he was convicted based on the testimony of Kiante
Watkins, one of the other three men involved in the offense, and the State failed to
submit sufficient evidence to corroborate Watkins’s version of what occurred. To
support a conviction based on the testimony of an accomplice, there must be
corroborating evidence that tends to connect the defendant with the offense. TEX.
CODE CRIM. PROC. ANN. art. 38.14. Corroborating evidence is not enough if it
merely shows the offense was committed. Id. But even apparently insignificant
incriminating circumstances, including confirming a mere detail, may provide
sufficient corroboration. Medrano v. State, 421 S.W.3d 869, 883 (Tex. App.—
–30–
Dallas 2014, pet. ref’d). We look at the facts and circumstances of each case and
consider the combined force of all the non-accomplice evidence that tends to connect
the accused to the offense. Id. While the defendant’s presence at the scene of the
crime is insufficient by itself to corroborate accomplice testimony, proof that the
defendant was at or near the scene, when coupled with other suspicious
circumstances, may be sufficient corroboration to support a conviction. Malone v.
State, 253 S.W.3d 253, 257 (Tex. Crim. App. 2008).
The location data generated by appellant’s phone showed he was at
Giddings’s house when Giddings was murdered. It also showed that he visited
Giddings’s neighborhood twenty days before the offense. Giddings’s security
camera recorded four men entering the house. Appellant’s phone records
demonstrate that, in the weeks leading up to and including the day of the offense, he
was communicating with the other three men seen in the security camera footage
when the offense occurred. His phone records also show he made a call from the
church parking lot across from Giddings’s house immediately before Giddings and
Dickerson were attacked. Appellant’s internet search history includes searches for
news about the offense and indicates appellant began reading a Dallas Police
Department blog, including a section on unsolved 2018 homicides, immediately
after the offense was committed.
Watkins testified that all four men arrived and fled from the scene in
appellant’s red Jeep Cherokee. Church surveillance video showed a Jeep Cherokee
–31–
leaving the church parking lot after the offense. The State submitted documentation
into evidence showing appellant purchased a red Jeep Cherokee approximately one
month before the murder. A few days after the offense, appellant was searching
online about how to reset the oil sensor for a Jeep Cherokee, indicating he still owned
the vehicle.
Although appellant attempts to discredit some of this evidence and Watkins’s
testimony in general, it was solely within the province of the jury to weigh the
evidence and evaluate credibility. See Dewberry v. State, 4 S.W.3d 735, 740 (Tex.
Crim. App. 1999). We conclude there was sufficient evidence tending to connect
appellant to the offense and to corroborate Watkins’s testimony. See Malone, 253
S.W.3d at 257. We resolve appellant’s fifth issue against him.
VI. Conspiracy Liability Instruction
In his final issue, appellant contends the trial court erred in instructing the jury
on conspirator liability. The challenged instruction, which tracks the language of
section 7.02(b) of the penal code, stated,
If, in the attempt to carry out a conspiracy to commit one felony,
another felony is committed by one of the conspirators, all conspirators
are guilty of the felony actually committed, though having no intent to
commit it, if the offense was committed in furtherance of the unlawful
purpose and was one that should have been anticipated as a result of the
carrying out of the conspiracy.
See TEX. PENAL CODE ANN. § 7.02(b). Appellant argues this instruction allowed the
jury to convict him without finding the specific intent required for the offense. He
further argues the instruction allowed the jury “to convict the non-shooter as a party
–32–
and similarly impose the death penalty” without being required to find the non-
shooter intended to cause the death or that the shooter intended to cause the death.
Appellant is mistaken.
It has long been the law in Texas that a person can be convicted of capital
murder as a party to the offense without having had the intent to commit the murder.
Ex parte Martinez, 330 S.W.3d 891, 901 (Tex. Crim. App. 2011); Johnson v. State,
853 S.W.2d 527, 535 (Tex. Crim. App. 1992). All the State was required to prove
in this case was that (1) appellant conspired with the group to commit robbery,
(2) the murder occurred in furtherance of the robbery, and (3) the murder should
have been anticipated as a result of carrying out the robbery. Martinez, 330 S.W.3d
at 902. The court’s charge properly instructed the jury on these elements.
Contrary to appellant’s suggestion that a finding of guilt based on conspiracy
liability allows a defendant to be sentenced to death without a finding of intent, the
code of criminal procedure requires additional findings before a defendant who was
found guilty as a party to capital murder can be subjected to the death penalty. If a
defendant is found guilty of a capital offense for which the State seeks the death
penalty, the trial court must conduct a separate sentencing proceeding to determine
whether the defendant will be sentenced to death or life in prison without parole.
TEX. CODE CRIM. PROC. ANN. art. 37.071, § 2(a). In cases where the jury charge
allowed the jury to find the defendant guilty as a party at the guilt/innocence stage,
the jury must be asked in the separate sentencing proceeding “whether the defendant
–33–
actually caused the death of the deceased or did not actually cause the death of the
deceased but intended to kill the deceased or another or anticipated that human life
would be taken.” Id. § 2(b)(2). Regardless, the State in this case did not seek the
death penalty and, therefore, appellees arguments are unavailing. See Cienfuegos v.
State, 113 S.W.3d 481, 495 (Tex. App.—Houston [1st Dist.] 2003, pet. ref’d).
We conclude the trial court’s charge to the jury on conspiracy liability was
proper. We overrule appellant’s sixth issue.
Based on the foregoing, we affirm the trial court’s judgment.
/Amanda L. Reichek/
AMANDA L. REICHEK
JUSTICE
Publish
TEX. R. APP. P. 47.2(b)
210855F.P05
–34–
S
Court of Appeals
Fifth District of Texas at Dallas
JUDGMENT
AARON RAYSHAN WELLS, On Appeal from the Criminal District
Appellant Court No. 4, Dallas County, Texas
Trial Court Cause No. F19-75986.
No. 05-21-00855-CR V. Opinion delivered by Justice
Reichek. Justice Goldstein
THE STATE OF TEXAS, Appellee participating.
Based on the Court’s opinion of this date, the judgment of the trial court is
AFFIRMED.
Judgment entered August 23, 2023
–35–