OPINION
FELIPE REYNA, Justice.
The State appeals an order granting Earl Cody Rudd’s motion to suppress evidence related to his arrest for driving while intoxicated and his motion to exclude the arresting officer’s testimony regarding Rudd’s performance on the horizontal gaze nystagmus (HGN) test. The State contends in three issues that the court erred by ruling: (1) “that a police officer may not lawfully question a witness to an accident unless the officer has reasonable suspicion to believe that person has also committed an offense”; (2) “that an officer has no probable cause to arrest a person the officer believes to be intoxicated unless the officer personally witnessed that person drinking and driving or operating a motor vehicle”; and (3) that the results of an HGN test are inadmissible even if the court finds that the officer was qualified and certified to perform that test. We will affirm in part and reverse and remand in part.
Background
DPS Trooper Kenneth Nolley responded to the scene of a single-vehicle accident at about 12:30 a.m. There were three people at the scene: the injured driver of a vehicle which had driven off the road into the woods; Rudd; and Nicole Stroope, who was the first to discover the accident. There were also three vehicles: the wrecked vehicle, Stroope’s car; and Rudd’s pickup. An ambulance transported the injured driver to an open area where he could be taken by helicopter to a hospital. Rudd rode in the ambulance to the helicopter landing site.
Stroope apparently followed the ambulance and then drove Rudd back to the accident scene.
Stroope told Nolley that she drove upon the accident scene after leaving Rudd’s home. When she arrived at the scene, she called 9-1-1 and Rudd at the injured driver’s request. At some point, Nolley’s partner mentioned to Nolley that Rudd had an odor of an alchoholic beverage about him. When Nolley asked Rudd how he came to the accident scene, Rudd said that “he responded” in his pickup after Stroope called him.
Nolley then asked Rudd to perform three field sobriety tests: the horizontal gaze nystagmus (HGN) test, the walk-and-turn test, and the one-leg-stand test. Nol-ley testified that he is trained and certified to perform these standardized tests and that he conducted these tests in accordance with his training. He observed four of a possible six clues of intoxication during the HGN test.
He testified that Rudd exhibited three of a possible eight clues of intoxication
during the walk-and-turn test because he: stepped out of the instructional position; used his arms for balance; and turned improperly. Rudd’s performance on the one-leg-stand test revealed no indicators of intoxication, although Nolley observed that Rudd “did not count out loud as I had instructed to him and demonstrated to him.” At some point, Rudd told Nolley that he had consumed about five or six alcoholic beverages during the course of the day.
On cross-examination, Rudd’s counsel highlighted the fact that Rudd apparently
did not have slurred speech or red, bloodshot eyes since Nolley did not mention these in his report and that Rudd was apparently able to provide accurate information about the injured driver. Nolley was then asked about reasons why an officer may choose not to make a video recording of the HGN test.
Q: Now, then, let’s talk about the fact that you did three tests. How many of them were videotaped?
A: Standardized field sobriety test, there were two that were videotaped.
Q: Is it possible — or could you have videotaped all three tests?
A: Yes, sir.
Q: Is it not true that the reason that you didn’t videotape the HGN is because you’ve been informed by prosecutors and other law enforcement officers to do it off camera, that way a defense attorney can’t attack the way you administered it?
A: I have been advised, sir.
Q: Is that why you did it?
A: No, sir.
Q: Why didn’t you videotape it if you could have?
A: Sir, I moved him to a position that was away from flashing lights that could interfere with one of the tests, sir.
Nolley was also cross-examined about his compliance with DPS videotaping policy,
about another trooper at the scene adjusting the camera during the field sobriety tests, and further about the decision to not videotape the HGN test.
Q: And you would agree that in this case you did not follow the DPS policy?
A: That’s incorrect, sir.
Q: Well, the videotape wasn’t on him when you were doing the HGN, was it?
A: It never states that it has to be on the suspect, sir. It has to be on, and it was.
Q: All right.
A: In accordance with the policy.
Q: Well, you — you would agree that when it came time for the one-leg-stand and the heel-to-toe, that not only did you move your car, but there were several times that you ran back to the ear and adjusted the camera?
A: I don’t believe I adjusted the camera, sir. I believe the trooper who was in the car adjusted the camera.
Q: And you all had a conversation about is that a good picture, does that show up?
A: Probably did, sir.
Q: Why didn’t you do it with the HGN?
A: Not required to, sir.
Q: Not required to. I thought earlier the reason you said you didn’t do it is because you were told by prosecutors and other troopers that that prevents a defense attorney to fig- ■ ure out whether or not you did it right?
A: I was told that, sir. But once again, I’m still not required to. And, still again, flashing lights could hinder the HGN test.
Q: Well, you can turn your flashing lights off, can’t you?
A: Some of them, sir. But I can’t control other people’s who — the ambulances on scene, fire trucks.
Q: You could have turned your car away around from them?
A: Could have done it off camera, and I did, sir.
Nolley further agreed that one of the purposes for the DPS videotaping policy is to allow supervisors to evaluate an officer’s performance and that, because he had not recorded Rudd’s HGN test, his supervisors could not evaluate whether he had properly conducted the test. He likewise conceded that neither defense counsel nor an expert could evaluate his conduct of the test.
Regarding the one-leg-stand test, Nolley conceded that he did not specifically instruct Rudd to count out loud even though the DPS manual provides that the test subject should be so instructed.
Regarding the walk-and-turn test, Nol-ley conceded that he left Rudd standing in the “instructional position” (heel touching toe) on three different occasions while Nol-ley adjusted the camera, turned off the flashers, and talked to the other trooper.
Q: Now, is that protocol under your— under your manual?
A: It doesn’t say that you can’t; it doesn’t say that you can.
Nolley was the only witness at the suppression hearing. At the conclusion of the hearing, the court ruled, “I’m going to grant the Motion to Suppress based on failure to have sufficient reasonable suspicion to conduct the standardized field sobriety test.” The court later signed an order containing detailed findings of fact and conclusions of law which are not separately numbered. The concluding paragraph of this order summarizes the court’s ruling as follows:
There was no reasonable suspicion to detain the Defendant at the time of his detention. The detention of the Defendant was unlawful under the circumstances presented by this case. The development of evidence subsequent to the unlawful detention of the Defendant was improper. Evidence obtained as a result of the unlawful detention of the Defendant is suppressed. The State failed to establish the admissibility of the HGN test and the observations or opinions related to the HGN test. The administration of the HGN test and the results therefrom are suppressed. The arrest without warrant of the Defendant was improper. The criteria for an arrest without a warrant is lacking in this case. Evidence obtained as a result of the arrest is suppressed.
Reasonable Suspicion
The State contends in its first issue that the court erred by “ruling that a police officer may not lawfully question a witness to an accident unless the officer has reasonable suspicion to believe that person has also committed an offense.” We construe this as a challenge to the court’s implicit conclusion that an officer must have reasonable suspicion to require a person to undergo field sobriety testing and the court’s express conclusion that Nolley did not have reasonable suspicion when he had Rudd perform field sobriety tests.
At a suppression hearing, the trial court is the sole judge of the credibility of witnesses and the weight to be given their testimony.
Wiede v. State,
214 S.W.3d 17, 24-25 (Tex.Crim.App.2007). Thus, the court may choose to believe or disbelieve any or all of a witness’s testimony.
Garza v. State,
213 S.W.3d 338, 346 (Tex.Crim.App.2007). “This Court is not at liberty to disturb any fact finding that is supported by the record.”
Id.
We view
the evidence in the light most favorable to the court’s ruling.
State v. Kelly,
204 S.W.3d 808, 818 (Tex.Crim.App.2006). When as here the court makes explicit findings of fact, we must determine whether the evidence supports those findings.
Id.
We then review the court’s legal ruling de novo unless the court’s findings (which are supported by the evidence) are also dispositive of the legal ruling.
Id.
“There are three recognized categories of interaction between the police and citizens: encounters, investigative detentions and arrests.”
State v. Larue,
28 S.W.3d 549, 553 n. 8 (Tex.Crim.App.2000) (quoting
Francis v. State,
922 S.W.2d 176, 178 (Tex.Crim.App.1996) (Baird, J., concurring and dissenting)). An encounter occurs when an officer approaches a person in public to ask questions.
See Florida v. Bostick,
501 U.S. 429, 434, 111 S.Ct. 2382, 2386, 115 L.Ed.2d 389 (1991);
Harper v. State,
217 S.W.3d 672, 674 (Tex.App.-Amarillo 2007, no pet.). An officer need not show any particular level of suspicion for such an encounter because the citizen is under no obligation to continue speaking with the officer.
Bostick,
501 U.S. at 434, 111 S.Ct. at 2386;
Larue,
28 S.W.3d at 553 n. 8;
Harper,
217 S.W.3d at 674. However, because an investigative detention and an arrest involve a seizure, an objective level of suspicion must be shown to justify the seizure.
Larue,
28 S.W.3d at 553 n. 8. For an investigative detention, the officer must show reasonable suspicion.
Id.; Harper,
217 S.W.3d at 675. For an arrest, probable cause is required.
Larue,
28 S.W.3d at 553 n. 8.
Generally, an officer’s questioning of a witness during an accident investigation is a consensual encounter.
See State v. Stevenson,
958 S.W.2d 824, 829 (Tex.Crim.App.1997);
Stoutner v. State,
36 S.W.3d 716, 719-20 (Tex.App.-Houston [1st Dist.] 2001, pet. refd);
see also Harper,
217 S.W.3d at 675 (officer’s questioning of person found passed out in vehicle in parking lot was an encounter). The answers to the officer’s questions and other observations by the officer may then provide reasonable suspicion to believe that the offense of DWI has occurred.
Stevenson,
958 S.W.2d at 829;
Stoutner,
36 S.W.3d at 719-20. If so, the encounter escalates to an investigatory detention during which the officer conducts field sobriety tests.
See Arthur v. State,
216 S.W.3d 50, 55 (Tex.App.-Fort Worth 2007, no pet.);
Reynolds v. State,
163 S.W.3d 808, 810-11 (Tex.App.-Amarillo 2005),
aff'd,
204 S.W.3d 386 (Tex.Crim.App.2006);
Stoutner,
36 S.W.3d at 719-20. The results of the sobriety testing may then lead to probable cause for an arrest.
See Rodriguez v. State,
191 S.W.3d 428, 444-45 (Tex.App.Corpus Christi 2006, pet. refd);
see also State v. Kurtz,
152 S.W.3d 72, 85 (Tex.Crim.App.2004) (Holcomb, J., dissenting) (citing
Stone v. State,
703 S.W.2d 652, 654-55 (Tex.Crim.App.1986)).
Here, the State is correct to say that Trooper Nolley did not need to have even reasonable suspicion to talk with Rudd at the accident scene and ask questions about the accident.
See Stevenson,
958 S.W.2d at 829;
Harper,
217 S.W.3d at 675;
Stoutner,
36 S.W.3d at 719-20. However, the State’s contentions are incorrect insofar as the State is asserting that an officer may require a person to undergo field sobriety tests without reasonable sus
picion that the person has committed an intoxication offense.
See Arthur,
216 S.W.3d at 55;
Reynolds,
163 S.W.3d at 810-11;
Stoutner,
36 S.W.3d at 719-20. Thus, the trial court correctly required a showing of reasonable suspicion to justify Rudd’s detention for field sobriety tests. We now examine the court’s conclusion that the State failed to establish the requisite reasonable suspicion.
“ ‘Reasonable suspicion’ exists if the officer has specific articulable facts that, when combined with rational inferences from those facts, would lead him to reasonably suspect that a particular person has engaged or is (or soon will be) engaging in criminal activity.”
Garcia v. State,
43 S.W.3d 527, 530 (Tex.Crim.App.2001). Reasonable suspicion is determined from the totality of the circumstances.
Id.
Although the court was free to disbelieve Trooper Nolley’s testimony, the State suggests that the court’s findings do not indicate that it disbelieved his testimony but rather that the court believed he did not identify sufficient articulable facts to justify the detention. Rudd does not agree with the State’s view and contends that court’s findings indicate that the court “did not believe much, if any, of the testimony provided by Nolley.”
The court’s findings do not provide much clarity in this regard. For example, the court found that Nolley “estimated” his time of arrival to be about fifteen minutes after Stroope called 9-1-1 and that Rudd returned from the helicopter landing area about one hour later. But there is nothing in the court’s findings to indicate that the court rejected these “estimates.”
Cf. State v. Cullen,
227 S.W.3d 278, 281 (Tex.App.-San Antonio 2007, pet. refd) (express finding that officer’s testimony “is not credible”);
Duhig v. State,
171 S.W.3d 631, 638 (Tex.App.-Houston [14th Dist.] 2005, pet. ref d) (express findings that officer was credible but defense witnesses were not). Our review of the court’s findings of fact and conclusions of law leads us to conclude that the court accepted Nolley’s testimony as credible (with the possible exception of his testimony regarding the field sobriety tests) but determined that the facts testified to by Nolley did not establish reasonable suspicion to detain Rudd for further investigation.
The court found there was no reasonable suspicion because: Nolley did not see Rudd driving his pickup; there was “no clear evidence” regarding if or when Rudd drove the pickup; Nolley did not observe any indications of intoxication before the other officer mentioned that Rudd smelled of alcohol; there was no evidence of Rudd’s conduct when he accompanied the victim to the helicopter landing site; and there was no evidence that Rudd intended or attempted to drive his pickup at any time after Nolley arrived at the scene.
We begin with the court’s finding that there was “no clear evidence” regarding if or when Rudd drove the pickup. Viewing the evidence in the light most favorable to this finding, the finding is simply not supported by the record.
See Kelly,
204 S.W.3d at 818. The court found that Stroope called 9-1-1 at about 12:15 a.m.
and that Nolley arrived at the scene about fifteen minutes later. From the testimony, Stroope called 9-1-1 and Rudd promptly after her arrival on the scene, and Rudd arrived at the scene before Nol-ley.
Regarding the question of “if’ Rudd drove the pickup, Nolley testified that there were three vehicles and three people at the scene when he arrived. Stroope
told Nolley that she called Rudd at the victim’s request. Rudd told Nolley that he “responded” to the scene when he received the call. Nolley “believed” that Rudd said that he had “responded” in the pickup.
It would not be reasonable to infer that Rudd was already present when Stroope came upon the accident scene in light of the victim’s request that Stroope call Rudd. Nor would it be reasonable to infer that Rudd’s pickup had been fortuitously parked at the accident scene earlier in the night. Therefore, the only reasonable inference to be drawn from the testimony is that Rudd drove to the accident scene in his pickup sometime between approximately 12:15 and 12:80 that morning.
The fact that Nolley did not personally observe Rudd driving the pickup is irrelevant. Article 14.08(a)(1) of the Code of Criminal Procedure provides in pertinent part that an officer may arrest a person found in a suspicious place under circumstances reasonably showing that he committed a violation of any of the intoxication offenses in Chapter 42 of the Penal Code.
See
Tex.Code Grim. PROC. Ann. art. 14.03(a)(1) (Vernon Supp.2007). Thus, an officer may have probable cause to arrest a person for DWI if the officer finds the arrestee in circumstances indicating that the arrestee committed this offense, even though the officer did not see the arrestee driving a vehicle.
See, e.g., Dyar v. State,
125 S.W.3d 460, 468 (Tex.Crim.App.2003);
Layland v. State,
144 S.W.3d 647, 650 (Tex.App.-Beaumont 2004, no pet.).
Reasonable suspicion is determined from the
totality of
the circumstances.
Garcia,
43 S.W.3d at 530. Here, the totality of relevant circumstances known to Trooper Nolley at the time he detained Rudd for field sobriety testing were: (1) Rudd had recently been driving his pickup; (2) Rudd had the odor of an alcoholic beverage about his person; and (3) Rudd told Nolley that he had consumed several alcoholic beverages earlier in the day. Thus, Nolley identified specific, articulable facts which provided reasonable suspicion to believe Rudd may have been driving his pickup while intoxicated.
See Stoutner,
36 S.W.3d at 720.
Accordingly, we sustain the State’s first issue.
Admissibility of HGN
The State contends in its third issue that the court erred by sustaining Rudd’s motion to exclude the HGN test results in view of the court’s finding that Trooper Nolley was qualified and certified to perform that test.
Testimony regarding HGN is scientific evidence, and its admissibility is governed by Rule of Evidence 702.
See Emerson v. State,
880 S.W.2d 759, 763 (Tex.Crim.App.1994);
Plouff v. State,
192 S.W.3d 213, 218 (Tex.App.-Houston [14th Dist.] 2006, no pet.);
Ellis v. State,
86 S.W.3d 759, 760 (Tex.App.-Waco 2002, pet. refd). Challenges to the admissibility of evidence under Rule 702 may be raised at a pretrial hearing, and the State may appeal an adverse ruling from such a hearing if the effect of that ruling is to “eviscerate the State’s ability to prove its case.”
State v. Medrano,
67 S.W.3d 892, 896 (Tex.Crim.App.2002). Thus, in a later proceeding the Court in
Medrano
addressed the State’s appeal of an adverse pretrial ruling on the admissibility of hypnotically enhanced testimony.
See State v. Medrano,
127 S.W.3d 781, 782 (Tex.Crim.App.2004).
Relying on an earlier version of Rule 702,
the Court of Criminal Appeals
explained in
Emerson
that scientific evidence must be helpful to the trier of fact to be admissible under Rule 702, which among other things requires that the evidence be reliable.
Emerson,
880 S.W.2d at 763;
accord Plouff,
192 S.W.3d at 218;
Ellis,
86 S.W.3d at 760. To be reliable, evidence regarding HGN (and scientific evidence in general) must satisfy three criteria: “1) the underlying scientific theory must be valid; 2) the technique applying the theory must be valid; and 3) the technique must have been applied properly on the occasion in question.”
Emerson,
880 S.W.2d at 763 (citing
Kelly v. State,
824 S.W.2d 568, 573 (Tex.Crim.App.1992));
accord Plouff,
192 S.W.3d at 218;
Ellis,
86 S.W.3d at 760. “The HGN test therefore must satisfy the three requirements set forth in
Kelly
in order to be admissible under Rule 702.”
Emerson,
880 S.W.2d at 764.
True, most of the focus in
Emerson
was on the first two criteria, namely, whether “the theory underlying the HGN test and the technique employed in its administration are both reliable.”
Emerson,
880 S.W.2d at 764. Accordingly, the Court devoted four pages of its opinion to a general examination of these two criteria, including a review of relevant literature and a survey of how other states address the admissibility of HGN.
Id.
at 764-68. The Court then applied the
three
criteria of
Kelly
to the facts of the case.
Id.
at 768-69. With regard to the third criteria, the Court concluded that the officer had followed the proper technique.
Id.
at 769.
Here, Rudd did not challenge and the trial court did not reject the reliability of HGN theory or of the recommended technique. Rather, the court concluded, “The testimony and record does [sic] not establish that the tests were administered properly and/or that the results from the tests are reliable.”
See Compton v. State,
120 S.W.3d 375, 377 (Tex.App.-Texarkana 2003, pet. ref'd) (“Under the
Emerson
three-part reliability test, even if a test’s underlying scientific theory and technique applying the theory are valid, improper application of the technique would render the test unreliable.”). Thus, the court focused on the third of the
Kelly
criteria. Although Trooper Nolley testified that he administered the HGN test in accordance with the recommended procedures, the court found his credibility to be lacking on this issue because of his failure to have Rudd perform the HGN test on video.
See Garza,
213 S.W.3d at 346 (court may choose to believe or disbelieve any or all of a witness’s testimony).
The State cites Compton for the proposition that any complaint regarding Nolley’s failure to videotape the HGN test and any “possible variation” on his part from the recommended procedure would affect only the weight to be given the HGN evidence and not its admissibility.
See Compton,
120 S.W.3d at 378-79. We disagree. In Compton, the court addressed what it described as “negligible” variances between the procedures employed by the arresting officer and the recommended procedures.
Id.
Thus, as the Fourteenth Court of Appeals has observed, Compton stands only for the proposition that “[s]light variations in the administration of the HGN test do not render the evidence inadmissible or unreliable, but may affect the weight to be given the testimony.”
Plouff,
192 S.W.3d at 219.
The court found that the State failed to prove that Trooper Nolley properly administered the HGN test to Rudd. The court was within its discretion to do so.
See
Garza,
213 S.W.3d at 346. Based on the State’s failure to prove the third requirement of
Kelly
for the admissibility of scientific evidence, the court did not abuse its discretion by excluding this evidence.
See Emerson,
880 S.W.2d at 764;
Compton,
120 S.W.3d at 377. Therefore, we overrule the State’s third issue.
Conclusion
The trial court erred m its conclusion that Trooper Nolley did not have reasonable suspicion to detain Rudd for field sobriety testing. Thus, the court erred by excluding evidence of field sobriety testing on the basis that it was obtained because of an invalid detention. However, the court did not abuse its discretion by excluding evidence of the HGN test because of the State’s failure to prove that Trooper Nolley had properly administered the test. Because the court did not address the issue of probable cause, we do not reach the State’s second issue.
See
Tex.R.App. P. 47.1.
We affirm that part of the trial court’s order excluding Trooper Nolley’s testimony regarding the HGN test. We reverse the remainder of the court’s order and remand this cause to the trial court for further proceedings consistent with this opinion.
Justice VANCE dissents from the judgment with a note.