The state appeals a pretrial order in these consolidated criminal cases that suppressed evidence discovered during a search of a residence pursuant to a search warrant. After discovering marijuana growing in the residence, the state charged each of the defendants with unlawful possession, delivery, and manufacture of marijuana. Defendants moved to suppress evidence derived from the search, and the trial court granted defendants’ motions by order. The state appeals that order pursuant to ORS 138.060. For the reasons set forth below, we reverse and remand.
On review of a trial court’s order suppressing evidence from a search conducted pursuant to a search warrant, we must determine whether, based on the facts in the affidavit for the search warrant, “a neutral and detached magistrate could conclude (1) that there is reason to believe that the facts stated are true; and (2) that the facts and circumstances disclosed by the affidavit are sufficient to establish probable cause to justify the search requested.” State v. Castilleja, 345 Or 255, 265,192 P3d 1283 (2008). That standard applies both to the trial court’s and our review of the issuing magistrate’s decision to issue the search warrant. Id. Thus, we do not defer to the trial court’s findings, because we are
“in the same position as was the trial court to evaluate the sufficiency of the facts alleged in the affidavit, the reasonableness of any inferences involved in resolving the legal question presented by the probable cause determination, and, ultimately, the existence of probable cause to support the warrant. No appellate court deference to the trial court’s findings or conclusions [is] appropriate or warranted.”
Id. (footnote omitted).
On July 19, 2010, Officer Shields of the Westside Interagency Narcotics (WIN) team obtained a warrant to search a residence at 3310 SW 173rd Avenue in Beaverton, occupied by defendant Melanie Orr. Shields’s affidavit in support of the warrant recited the following facts. In February 2010, an anonymous caller called the Beaverton police department to report a possible marijuana-growing operation at *58Orr’s residence. Shields contacted Portland General Electric (PGE) and determined that Orr had been the power subscriber at the residence since 2005 and that the power usage at the address had increased dramatically in June 2008 and had remained very high. Shields then contacted the Oregon Medical Marijuana Program (OMMP) and determined that Orr was a registered medical-marijuana user who was authorized to grow marijuana (up to 6 mature plants and 18 seedling plants) at that address.
In May 2010, Shields received calls from two informants who wished to keep their identities confidential. The first of the informants reported that there was sporadic vehicular and foot traffic by people who would visit the residence for fewer than 10 minutes. The informant said that the people would arrive at the residence carrying nothing and leave carrying gym or duffel bags. The informant indicated that that activity followed a cycle in which the short-duration visits would persist for roughly a week, decrease for three to four weeks, and then increase again. The informant also reported smelling an odor of marijuana at the residence.
The second informant reported seeing a young man from the neighborhood visit the residence several times and leave shortly thereafter. The informant also reported repeatedly seeing a pizza-delivery vehicle arrive at the residence and the driver, a young man, go in carrying nothing and later emerge carrying a duffel bag. Finally, the informant expressed the belief that marijuana was being sold from the residence.
In May and June of 2010, Shields received detailed records from PGE on power usage at the residence. Based on power records attached as an exhibit to his affidavit, Shields provided calculations indicating that the increased power consumption was consistent with the use of seven metal halide lights of the type used in indoor marijuana-growing operations, which would support the growth of somewhere between 28 and 140 mature marijuana plants.1 Also, Shields went to the residence and noted that its only ground-level *59window was covered from the inside, which was consistent with a marijuana-growing operation being conducted inside.
On July 6, 2010, Shields and WIN officers established surveillance of the residence. Shields watched a young man come to the house, stay for approximately 10 minutes, then leave with his left hand in the pocket of his shorts. Several minutes later, another officer saw what he believed to be a hand-to-hand transaction near the residence between a person in a dark blue Chevrolet TrailBlazer and a person in a gold Honda Civic. The officer lost sight of both vehicles approximately one block from the residence, but Shields saw the gold Honda arrive at the residence about two minutes later. The TrailBlazer then drove by the residence and parked roughly a block away. A man got out of the Honda, went into the residence, and came out five minutes later. The man returned to the Honda and drove to the parked TrailBlazer. Shields and another officer watched what appeared to be a hand-to-hand transaction between the driver of the TrailBlazer and a person in the Honda.
Shields followed the TrailBlazer, saw its driver commit traffic infractions, and had another officer stop the vehicle for the infractions. During the traffic stop, the driver, Slover, admitted possessing approximately one half ounce of marijuana. She said that she had just purchased it from a man in a small tan Honda at a location that was consistent with the officers’ observation of the hand-to-hand transaction that had occurred several minutes earlier. Shields further averred:
“Based on my training and experience and the observation of the [other officers], the initial contact between the Trailblazer and Honda Civic was an exchange of money from Slover to the passenger in the Honda Civic. The Honda Civic proceeded to 3310 SW 173rd Avenue and the driver of the Honda Civic entered 3310 SW 173rd Avenue and purchased illegal narcotics from inside the residence. The driver of the Honda Civic exited 3310 SW 173rd Avenue and drove from that location and located the Trailblazer a block away from 3310 SW 173rd Avenue where there was a hand to hand exchange between the front passenger of the Honda Civic and Slover. Based on my training and experience, this would be consistent with an exchange of illegal narcotics from the occupants of the Honda Civic to Slover.”
*60Shields also recited details of his training and experience in investigating drug-related crimes, described the types of evidence expected to be located at sites where drugs are manufactured or delivered, and opined that (1) frequent visits of a short duration at a location can be indicative of drug distribution, (2) indoor marijuana-growing operations typically are conducted on a year-round basis, and (3) people who grow more marijuana than is allowed under the OMMP often distribute their excess marijuana for consideration. As noted, a search warrant issued for a search of the residence at 3310 SW 173rd Avenue, and defendants moved to suppress the evidence discovered during the search of the residence.
The trial court held a hearing on the suppression motion, during which defendants focused on what they believed to be a crucial weakness in the affidavit in support of the warrant, viz., that the information from the two unnamed informants who had contacted Shields in May 2010 had not been verified or established to be reliable. The trial court granted the motion, citing several difficulties with Shields’s affidavit. The court agreed with defendants that the information from the informants had not been verified in any meaningful way and that their information therefore did not add any weight to the affidavit. The court also noted that, because the residence was an authorized marijuana-growing site under the OMMP, the information in the affidavit indicating that marijuana was being grown at the residence did not provide probable cause to believe that unlawful activity was occurring at the residence. Finally, the court discounted the power-consumption data, stating that it was not clear “that the information takes into account all the other factors that might have affected the amount of power consumption, and * * * there’s no comparison to another similar place.”
On appeal, the state argues that the trial court erred in granting the suppression motion, focusing in particular on the information in the affidavit about the July 6 drug transaction that was linked to the residence. The state also argues that the issuing magistrate could have inferred from the information provided by the unnamed informants *61that they had made first-hand observations of the activities that they had described to Shields, and, because the informants were known by name to Shields, their veracity was established because they would be subject to criminal penalties if they made false reports.
We need not resolve whether the trial court correctly concluded that Shields had failed to present facts in his affidavit sufficient to support a determination that the information given to him by the unnamed informants was reliable, because we conclude that the remaining information in the affidavit is sufficient to provide probable cause for the search. We begin by noting that, in exercising our review function, we
“view the predicate affidavit in a ‘commonsense, nontechnical and realistic fashion,’ with ‘doubtful cases * * * to be resolved by deferring to an issuing magistrate’s determination of probable cause.’ State v. Wilson, 178 Or App 163,167, 35 P3d 1111 (2001) (internal quotation marks omitted). That deferential standard comports with ‘the preference for warranted searches over those conducted without prior judicial authorization.’ Id.”
State v. Duarte/Knull-Dunagan, 237 Or App 13, 21-22, 238 P3d 411, rev den, 349 Or 370 (2010) (ellipses in Duarte).
In Duarte, as in this case, police officers received information from callers about an alleged marijuana-growing operation, but they arguably failed to sufficiently establish the informants’ reliability. Id. at 22. We held that “the issuing magistrate could reasonably have concluded that the totality of the facts set out in the affidavit, separate from the informant’s allegations, sufficiently corroborated the informant’s statements that there was a hydroponic marijuana [-] growing operation” in the target residence. Id. Specifically, we noted the affiant’s averments about his training and experience with growing operations of that kind; increased water consumption over that of the previous owner, as well as electrical consumption that was substantially greater than that for comparable properties; visual confirmation of certain structural modifications at the residence; and the defendant’s prior arrest and conviction for manufacture and *62delivery of a controlled substance. Id. at 22-23. Based on those facts, we concluded that the issuing magistrate “could properly determine that it was probable that the items identified in the warrant application” were subject to seizure. Id. at 25.
This case is similar. As noted, Shields was not able to confirm the specific details that had led the unnamed informants to believe that there was an unlawful marijuana-growing operation at defendants’ residence, viz., the smell of marijuana, the heavy vehicular and foot traffic, the people arriving empty-handed and leaving with bags. However, Shields developed a significant amount of other information that tended to show that unlawful drug activity was occurring at the residence.
It is not seriously disputed that Shields had ample information to establish that marijuana was being grown at the residence. The location was registered with OMMP as a marijuana-cultivation site, it had a window covered in a manner that is associated with indoor marijuana cultivation, and power records supported an inference that grow lights were being used to grow marijuana in the residence.
The question is whether the affidavit could properly support a determination that there was probable cause that unlawful activity related to marijuana cultivation was occurring. In particular, we note several significant aspects of Shields’s training and experience. First, Shields stated that, in his experience, people who are authorized to grow marijuana under the OMMP and grow more than the allowed amount under the program often become involved in the unlawful distribution of marijuana. Second, he described in detail his and other officers’ observations of the events on July 6 that ultimately led Slover to admit that she had purchased marijuana. And, based on his training and experience, Shields concluded that what the officers had observed was consistent with Slover making contact with an occupant of the Honda and providing him with money, after which the Honda went to defendants’ address and an occupant went inside, purchased marijuana, and delivered it to Slover at a nearby location. Slover’s admissions were entirely consistent with Shields’ understanding of what the officers had *63witnessed. Those facts reasonably corroborate a conclusion that an occupant of the Honda had acquired marijuana at defendants’ residence and thereafter sold it.
Finally, we turn to the significance of the power consumption records that Shields included as an exhibit to his affidavit, as well as the materials in the affidavit itself concerning Shields’s calculations. Because that information is voluminous, we provide a very brief summary of it. Shields detailed his extensive training and experience with marijuana-growing operations and described typical marijuana-growing operations as involving three stages: propagation, vegetation, and budding. In indoor-growing operations, each stage involves exposing the plants to differing amounts of light. Metal halide lights are not used in the propagation stage, but they are used in the later two stages. Each metal halide light commonly used for such operations draws an average of 12 or 18 kilowatts of electricity per day, depending on the light cycle. Each light can support the growth of between 4 and 20 marijuana plants. Between February 2007 and June 2008, the daily electrical consumption at the residence averaged 75.56 kilowatts of power per day. Between July 2008 and June 2010, the daily electrical consumption at the residence averaged 198.56 kilowatts per day. Shields concluded that the increase in power consumption could be due to the use of seven metal halide lights in the residence. Finally, Shields concluded that the use of seven metal halide lights was consistent with the growth of a great many more mature marijuana plants than defendant Orr was authorized to grow under the OMMP.
Defendants argue that the power records simply cannot support an inference that a large number of metal halide lights were being used at the residence, noting that Shields’s calculations attributed all of the increase in power consumption to the lights and did not eliminate the possibility that other items in the residence were responsible for the increase in power consumption. However, just as “an observation made by police that is consistent with criminal conduct does not have to eliminate any possibility of an innocent explanation to provide probable cause,” State v. Foster, *64350 Or 161, 173, 252 P3d 292 (2011), a calculation by a police officer, based on the officer’s significant amount of training and experience with marijuana-growing operations, about the probable size of a marijuana-growing operation similarly may contribute to a finding of probable cause.
Defendants also cite State v. Milks/Sales, 127 Or App 397, 872 P2d 988 (1994), for the proposition that the power-consumption information and records here do not contribute to probable cause. We disagree. In Milks/Sales, the affiant compared the defendants’ power usage to the power usage of a prior tenant several years earlier and concluded that the increased electrical usage could be accounted for by “the use of approximately two or three grow lights.” Id. at 400. We concluded that the affidavit in that case (which was supported by virtually nothing besides the power-consumption data) did not establish probable cause, noting that comparison with the prior tenant’s power usage was not particularly significant because “the magistrate had no way of knowing whether the current tenant’s power-consumption was abnormally high, the prior tenant’s was abnormally low or whether both were within the range of normal use.” Id. at 403. We went on to suggest that the following type of data might be of use:
“[A] comparison made by a utility company official who states that the defendant’s electrical usage was excessive, a comparison to a number of similar structures used for the same purpose over the same period of time, a comparison to estimated figures, from a utility company brochure that were previously found to be accurate, and a comparison to normal residential usage.”
Id. at 403 n 6 (citations and internal quotation marks omitted). That list, however, is hardly exhaustive, and the result in Milks /Sales was based in large part on the fact that the power-consumption data were virtually all of the information that the affiant had presented to establish probable cause. See, e.g., State v. Johnson, 186 Or App 186, 191-92, 62 P3d 861 (2003) (power consumption data was considered along with other information in determining probable cause *65existed for issuance of a warrant); State v. Poppe, 131 Or App 14, 23, 883 P2d 905, rev den, 320 Or 492 (1994) (same).
Here, in contrast to Milks!Sales, Shields had power-consumption records for the residence spanning a three-year period during which Orr was the subscriber. Thus, it is not a question of comparing data from a former tenant to a later one. Moreover, when Shields contacted PGE, the PGE representative with whom he spoke told him “that the power usage for this address dramatically increased from June 2008 through September 2008 and maintained higher levels of power usage to the present time.” The specific data that Shields later obtained from PGE through administrative subpoena were consistent with that statement. That is, in this case, we have a large increase in power consumption by the same tenant, described by a PGE representative as a “dramatic” increase.
In addition, and unlike in Milks!Sales, we also have other significant information that contributes to the analysis — in particular, that Shields had determined that the residence was registered with the OMMP as a marijuana-growing site and had a covered window consistent with a marijuana-growing site. Thus, Shields’s conclusion that marijuana was being grown at the residence is supported by more than just the power records.
The ultimate question, though, is whether there was probable cause to believe that unlawful marijuana cultivation and sales were occurrings there. And, again, Shields had more than just the power records to support the conclusion that there were. Not only did the power-consumption records support an inference that marijuana was being grown there, but, given Shields’s training and experience, the records supported an inference that a large number of marijuana plants were being grown there. In addition, the observations of the drug transaction on July 6 involving the Honda and Slover supported an inference that marijuana grown at the residence was being unlawfully distributed.
We thus conclude that the magistrate properly issued a warrant for the police to search the residence for *66items related to unlawful drug activity. The trial court erred in granting defendants’ suppression motion.
Reversed and remanded.
That aspect of the affidavit is described in more detail below. 259 Or App at 63.