Lead Opinion

J.M. Johnson, J.

¶1 This case concerns a marriage dissolution and foreign travel restrictions imposed as part of a parenting plan under RCW 26.09.191(3). The restrictions were imposed by the trial court based on evidence that the father made threats to abscond with his children to India.1 The father denies making such threats and claims the restrictions are not supported by the trial court’s findings. He further argues the court committed prejudicial error by allowing improper expert testimony regarding “risk factors” for child abduction. The Court of Appeals affirmed the trial court’s parenting plan and travel restrictions, concluding the admission of risk factor evidence was *29improper but not prejudicial. We affirm the Court of Appeals except for its conclusion that the trial court erred by admitting the expert testimony. We uphold the travel restrictions because the trial court’s findings are supported by substantial evidence and admission of the expert testimony was not an abuse of discretion.

Facts

¶2 Petitioner Brajesh Katare was born and raised in India. In 1989, he moved to Florida where he met respondent Lynette. The two were married in November 1995. Brajesh and Lynette relocated to Washington State in 1999 when Brajesh was hired by Microsoft. While in Washington, Brajesh and Lynette had two children: A.K. (born May 27, 2000) and R.K. (born Sept. 20, 2001).

¶3 In May 2002, Brajesh was offered employment in India for two years. Lynette strongly objected to moving to India with the children, fearing isolation and deleterious effects on the children’s health. Lynette claims Brajesh threatened to take the children to India without her, but Brajesh denies making such threats.

¶4 In July 2002, Brajesh traveled to India to make arrangements to move. While he was gone, Lynette filed for dissolution. Brajesh and Lynette eventually agreed to a temporary parenting plan allowing Brajesh biweekly supervised visits with the children. They also agreed to appoint Margo Waldroup, a parenting evaluator, to conduct an assessment and make a parenting plan recommendation to the court.

¶5 Waldroup recommended the children remain in Lynette’s custody. She noted that two witnesses corroborated the allegation that Brajesh had threatened to take the children to India. However, Waldroup could not predict with certainty whether Brajesh would actually attempt to abduct the children. Waldroup recommended that Brajesh have three days of visitation with the children each month. *30She recommended supervised visitations until the children’s passports were secured. She also suggested Brajesh’s passport, as well as those belonging to the children, be placed on a watch list. Based on Waldroup’s report, Brajesh moved to modify the temporary parenting plan to allow him unsupervised visitation. The court granted his motion subject to the requirement his attorney hold his passport during visitations.

Procedural History

1. Katare I

¶6 A five-day dissolution hearing was held in June 2003. Lynette asked the court to impose restrictions on Brajesh’s visitation time. Lynette testified that Brajesh on multiple occasions threatened to take the children to India without her. Brajesh allegedly told her she would have no recourse if he took the children to India and she would not “stand a chance” in the Indian court system. Verbatim Report of Proceedings (VRP) (June 16, 2003) at 36. During discovery, Brajesh requested copies of the children’s passports, Indian tourist visas, and immunization records, which Lynette claimed showed his intent to take the children to India. Lynette was especially concerned because India is not a signatory to the Hague Convention on the Civil Aspects of International Child Abduction (Hague Convention), which provides for mandatory summary proceedings in cases of international child abduction. The treaty provides a remedy only if both countries are signatories. Therefore, it would be especially difficult for Lynette to get the children back to the United States if they were taken to India.

¶7 Waldroup testified that although Lynette’s concerns were “justified” and “not out of proportion to the situation,” it was difficult to predict whether Brajesh would abduct the children. VRP (June 17, 2003) at 103. She testified that the court would have to decide whether the risk justified imposing restrictions.

*31¶8 Brajesh categorically denied making any threats. He represented he did not want to return to India and would never take the children away from their mother.

¶9 The trial court imposed restrictions in the parenting plan even though it found RCW 26.09.191(3) inapplicable.2 In its oral decision, the court stated:

I gave a long and careful consideration to the issue of the risk of abduction and confess today being concerned about this. I’m not persuaded, based on all the evidence presented ... that Mr. Katare presents a serious threat of abducting the children. Nonetheless, if I’m wrong on this the consequences are incredibly serious and I’m mindful about that. I’m going to impose some restrictions in the parenting plan that will be designed to address this issue, and I hope that everything that has been brought to this Court, which I think indicates that there is not a serious risk of abduction [,] turns out to be the truth.

VRP (July 7, 2003) at 10.

¶10 The court’s parenting plan allowed Brajesh three days with the children each month. Brajesh was prohibited from taking the children out of the country until they turned 18. Brajesh was also denied access to the children’s passports or birth certificates and was required to surrender his passport to a neutral party during visitation periods.

¶11 Brajesh filed a motion for reconsideration, challenging the visitation restrictions and passport controls. The court denied his motion. Brajesh appealed. He argued the trial court erred when it imposed restrictions without expressly finding factors justifying the restrictions under RCW 26.09.191(3).

*32¶12 The Court of Appeals held restrictions entered in a parenting plan pursuant to RCW 26.09.191(3) must be supported by an express finding that the parent’s conduct is adverse to the best interest of the child. In re Marriage of Katare, 125 Wn. App. 813, 826, 105 P.3d 44 (2004) (Katare I). While most of the restrictions imposed by the trial court were supported by the court’s findings, the court’s order stating that RCW 26.09.191(3) “ ‘does not apply’ ” created an ambiguity. Id. at 831 (quoting trial court clerk’s papers at 615). For that reason, the Court of Appeals remanded to clarify the legal basis for the foreign travel restrictions. Id.

2. Katare II

¶13 On remand, the trial court amended the parenting plan to list factors justifying the restrictions under RCW 26.09.191(3)(g). The amended parenting plan provided:

“Based on the evidence, including the testimony of expert witnesses, the husband appears to present no serious threat of abducting the children. Nonetheless, under the circumstances of this case, given the ages of the children, the parties’ backgrounds, ties to their families and communities, and history of parenting, and the fact that India is not a signator to the Hague Convention on International Child Abduction, the consequences of such an abduction are so irreversible as to warrant limitations on the husband’s residential time with the children. The risk of abduction is a factor justifying limitations under RCW 26.09.191(3)(g).”

In re Marriage of Katare, noted at 140 Wn. App. 1041, 2007 WL 2823311, at *2-3, 2007 Wash. App. LEXIS 2755, at *9-10 (Katare II) (“By basically restating its earlier findings as the justification for imposing limitations on Brajesh’s residential time with the children under RCW 26.09.191(3)(g), the trial court does not resolve the ambiguity and does not expressly address whether the evidence supports the limitations under RCW 26.09.191(3).”).

¶14 The Court of Appeals found this recitation deficient as it still contained the phrase “ ‘the husband appears to *33present no serious threat of abducting the children.’ ” 2007 WL 2823311, at *3, 2007 Wash. App. LEXIS 2755, at *10 (quoting parenting plan). The appellate court remanded a second time for clarification. 2007 WL 2823311, at *3, 2007 Wash. App. LEXIS 2755, at *10.

3. Katare III

¶15 In January 2009, the trial court conducted a two-day hearing to address whether the evidence supported the foreign travel restrictions and passport controls. The court heard testimony from Brajesh and Lynette and considered a number of hostile e-mail exchanges between the two.

¶16 Lynette also identified an expert witness, Michael C. Berry, an attorney with 17 years of experience with child abduction cases, to testify regarding risk factors for child abduction and the consequences of abduction to India. Brajesh filed a motion in limine to exclude Berry’s testimony. He argued Berry was not qualified as an expert and that risk factor evidence was inadmissible under Frye 3 The court allowed Berry to testify, noting his testimony would “assist [the court] in understanding the status of the literature on these topics.” VRP (Jan. 14, 2009) at 81.

¶17 Through Berry’s testimony, Lynette introduced literature citing “risk factors” for child abduction.4 Id. Berry observed that several of the risk factors applied to Brajesh, *34specifically: Brajesh had previously threatened to abscond with the children to India. He had strong ties to India and no family in the United States other than the children. Brajesh engaged in planning activities evidencing his intent to move to India, including selling a car and attempting to obtain the children’s passports and immunization records. He accused Lynette of lying and abuse. He plainly felt disenfranchised by what he called the “biased” legal system. Finally, there was a clear lack of cooperation between the parents.

¶18 The trial court entered detailed findings of fact and conclusions of law on second remand.5 The court eliminated its earlier finding that Brajesh appeared to pose no serious threat of abducting the children. Instead, the “extreme anger, abuse, unreasonableness, and poor judgment” that Brajesh demonstrated convinced the court “[t]he risk of abduction ha[d] not abated,” but had perhaps increased. Clerk’s Papers (CP) at 154. Accordingly, the court concluded it was in the best interest of the children to maintain the travel restrictions.

¶19 Brajesh appealed, arguing the trial court erred by maintaining the travel restrictions and denying his motion to exclude Berry’s expert testimony. The Court of Appeals found the restrictions were supported by substantial evidence. In re Marriage of Katare, noted at 159 Wn. App. 1017, 2011 WL 61847, at *10, 2011 Wash. App. LEXIS 65, at *28 (Katare III). It held the trial court abused its discretion by admitting Berry’s testimony because there was not an *35adequate foundation and Lynette did not establish that the risk factor evidence met the Frye standard. 2011 WL 61847, at *12, 2011 Wash. App. LEXIS 65, at *35. The Court of Appeals nonetheless found the error harmless because the trial court did not adopt Berry’s risk factor analysis as its own. 2011 WL 61847, at *12, 2011 Wash. App. LEXIS 65, at *35. Instead, the trial court had noted the restrictions were supported by Brajesh’s “testimony and conduct alone.” 2011 WL 61847, at *9, 2011 Wash. App. LEXIS 65, at *28.

¶20 Brajesh petitioned this court for review, which was granted. In re Marriage of Katare, 171 Wn.2d 1021, 257 P.3d 662 (2011).

Analysis

1. Standard of Review

¶21 A trial court’s parenting plan is reviewed for an abuse of discretion. In re Marriage of Littlefield, 133 Wn.2d 39, 46, 940 P.2d 1362 (1997). An abuse of discretion occurs when a decision is manifestly unreasonable or based on untenable grounds or untenable reasons. Id. at 46-47. The trial court’s findings of fact will be accepted as verities by the reviewing court so long as they are supported by substantial evidence. Ferree v. Doric Co., 62 Wn.2d 561, 568, 383 P.2d 900 (1963). Substantial evidence is that which is sufficient to persuade a fair-minded person of the truth of the matter asserted. King County v. Cent. Puget Sound Growth Mgmt. Hr’gs Bd., 142 Wn.2d 543, 561, 14 P.3d 133 (2000).

2. The Foreign Travel Restrictions Imposed by the Trial Court Are Supported by Substantial Evidence

¶22 A trial court wields broad discretion when fashioning a permanent parenting plan. In re Marriage of Kovacs, 121 Wn.2d 795, 801, 854 P.2d 629 (1993). The court’s discretion must be guided by several provisions of the Parenting Act of 1987, namely RCW 26.09.187(3) (enumer*36ating factors to be considered when constructing a parenting plan), RCW 26.09.184 (setting forth the objectives of the permanent parenting plan and the required provisions), RCW 26.09.002 (declaring the policy of the Parenting Act of 1987), and RCW 26.09.191 (setting forth factors which require or permit limitations upon a parent’s involvement with the child). Id. Relevant to this case, RCW 26.09.191(3)(g) allows the trial court to limit the terms of the parenting plan if it finds a parent’s conduct is “adverse to the best interests of the child.” Imposing such restrictions “require [s] more than the normal... hardships which predictably result from a dissolution of marriage.” Littlefield, 133 Wn.2d at 55.

¶23 Brajesh first contends foreign travel restrictions were improperly imposed because the trial court found only a “risk” of abduction. According to Brajesh, absent actual harmful conduct toward the child, restrictions cannot be levied. However, the trial court need not wait for actual harm to accrue before imposing restrictions on visitation. In re Marriage of Burrill, 113 Wn. App. 863, 872, 56 P.3d 993 (2002) (“evidence of actual damage is not required”). “Rather, the required showing is that a danger of . . . damage exists.” Id. Because the trial court found a danger of serious damage (abduction) here, restrictions were appropriate even though Brajesh had not yet attempted abduction. See also Lee v. Lee, 49 So. 3d 211, 215 (Ala. Civ. App. 2010) (“[A] number of cases in American jurisdictions recognize the propriety of [limited] visitation when the noncustodial parent is shown to pose a risk of abduction.” (citing Shady v. Shady, 858 N.E.2d 128, 143 (Ind. Ct. App. 2006); Moon v. Moon, 277 Ga. 375, 377, 589 S.E.2d 76, 79-80 (2003); Monette v. Hoff, 958 P.2d 434, 436 (Alaska 1998))).6

*37¶24 Brajesh next contends the Court of Appeals’ decision in this matter conflicts with In re Marriage of Wicklund, 84 Wn. App. 763, 932 P.2d 652 (1996), and In re Marriage of Watson, 132 Wn. App. 222, 130 P.3d 915 (2006). According to Brajesh, these cases establish that abduction must be likely before his visitation time may be limited. But Wicklund and Watson simply indicate that restrictions cannot be imposed for unfounded reasons.

¶25 In Wicklund, 84 Wn. App. at 769, the trial court restricted a father’s ability to display affection with his partner in front of his children because the children were having difficulty adjusting to his homosexuality. The Court of Appeals held the restrictions were an abuse of discretion because “[p]roblems with adjustment are the normal response to any breakup of a family. ... If the problem is adjustment, the remedy is counseling.” Id. at 771. Unlike adjustment issues, the threat of abduction goes well beyond “the normal response to any breakup of a family.” Therefore, Wicklund does not support Brajesh’s position.

¶26 In Watson, 132 Wn. App. at 227-28, the trial court imposed restrictions on a father’s visitation time based solely on the mother’s unfounded allegation that he had abused their daughter. The Court of Appeals reversed because “the unproven allegation of sexual abuse [did] not provide substantial evidence in support of the visitation restrictions” and the remaining evidence weighed in favor of the father. Id. at 233. In contrast, Lynette’s allegations have been corroborated. Two witnesses heard Brajesh’s threats and submitted sworn statements to that effect. The parenting evaluator found Lynette’s fears “justified.” Moreover, the trial court considered additional evidence of Brajesh’s conduct, including eyewitness testimony and e*38mail exchanges. After weighing all the available evidence, the court found Brajesh’s “pattern of abusive, controlling, punishing behavior put[ ] the children at risk of being used as the tools to continue this conduct.” CP at 156. Thus, substantial evidence — evidence sufficient to persuade a fair-minded person that Brajesh posed a risk of abduction— justified the passport restrictions under Watson.7

3. The Trial Court Did Not Abuse Its Discretion by Admitting Berry’s Expert Testimony Regarding Risk Factors for Parental Child Abduction

¶27 Generally, a party may introduce expert testimony as long as the expert is qualified, relies on generally accepted theories, and assists the trier of fact. ER 702. Determining the admissibility of expert evidence is largely within a trial court’s discretion. Philippides v. Bernard, 151 Wn.2d 376, 393, 88 P.3d 939 (2004). “ ‘[T]he exercise of [such discretion] will not be disturbed by an appellate court except for a very plain abuse thereof.’ ” Hill v. C&E Constr. Co., 59 Wn.2d 743, 746, 370 P.2d 255 (1962) (quoting Wilkins v. Knox, 142 Wash. 571, 577, 253 P. 797 (1927)).

¶28 Brajesh alleges Berry, the expert called by Lynette at the second remand hearing, was not qualified to testify as an expert on risk factors for child abduction or to attest to the consequences of abduction to India. An expert may not testify about information outside his area of expertise. Queen City Farms, Inc. v. Cent. Nat’l Ins. Co. of Omaha, 126 Wn.2d 50, 104, 882 P.2d 703, 891 P.2d 718 (1994). While Berry’s formal education was not related to child abduction, an expert may be qualified by experience alone. ER 702. Berry had 17 years of experience in the field of child abduction, during which he participated in related organi*39zations, attended numerous conferences, consulted with governmental entities, and testified as an expert in other abduction cases. Given the length and range of Berry’s experience, it was not an abuse of discretion for the court to have concluded that his testimony would be helpful.

¶29 Brajesh also argues Berry’s testimony lacked an adequate foundation because Berry had never been to India and had never interacted with Brajesh personally. Expert opinions lacking an adequate foundation should be excluded. Walker v. State, 121 Wn.2d 214, 218, 848 P.2d 721 (1993). But, an expert is not always required to personally perceive the subject of his or her analysis. ER 703 (“The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to the expert at or before the hearing.” (emphasis added)). That an expert’s testimony is not based on a personal evaluation of the subject goes to the testimony’s weight, not its admissibility. Since Berry’s testimony was based on information made known to him before the hearing, the court did not abuse its discretion by admitting his testimony without establishing his personal familiarity with Brajesh or India.

¶30 The Court of Appeals held the risk factor evidence was improperly admitted, analogizing the risk factors to propensity evidence. Katare III, 2011 WL 61847, at *12, 2011 Wash. App. LEXIS 65, at *33-35. But, deciding whether to impose restrictions based on a threat of future harm necessarily involves consideration of the parties’ past actions. By its terms, RCW 26.09.191(3) obligates a trial court to consider whether “[a] parent’s involvement or conduct may have an adverse effect on the child [ren]’s best interests.” (Emphasis added.) To make this determination, the court must engage in a form of risk assessment.

¶31 This court approved the use of risk assessments to predict the future dangerousness of sexually violent predators in In re Detention of Thorell, 149 Wn.2d 724, 72 P.3d 708 (2003). While recognizing that predictions of future *40dangerousness are necessarily prejudicial, we nonetheless held such testimony is admissible because the probative value is “high and directly relevant” to whether an offender should be civilly committed. Id. at 758. We explicitly stated such assessments are “not profile evidence.” Id. Similarly, the risk factor evidence at issue here was “directly relevant” to whether visitation restrictions were necessary — a determination that unavoidably involved prediction. Therefore, as in Thorell, the risk factor evidence in this case was not inadmissible “profile evidence,” but was properly admitted and utilized.8

4. The Trial Court Did Not Commit a Constitutional Violation by Considering the Risk Factor Evidence or by Imposing Reasonable Foreign Travel Restrictions as Part of a Parenting Plan

¶32 Brajesh’s principal arguments over the life of this case have shifted to take on constitutional overtones. In his petition for review, Brajesh cites no authority for his assertion that “this Court should ... declare that, whenever a Washington trial court relies on racial profiling evidence to impose restrictions, the usual deference to the trial court’s decisions will be subject to the strictest scrutiny....” Pet. for Review at 20. But, “ 'naked castings into the constitutional sea are not sufficient to command judicial consideration and discussion.’ ” In re Pers. Restraint of Williams, 111 Wn.2d 353, 365, 759 P.2d 436 (1988) (internal quotation marks omitted) (quoting In re Rosier, 105 Wn.2d *41606, 616, 717 P.2d 1353 (1986)). Brajesh’s quasi-constitutional arguments do not warrant an extensive discussion.

¶33 Firstly, the risk factors considered by the trial court cannot conceivably be regarded as “racial profiling evidence.” The vast majority of the factors considered had nothing to do with race or national origin, including (1) whether there has been a prior threat of abduction; (2) whether the parent has engaged in planning activities that could facilitate removal of the child from the jurisdiction; (3) whether the parent has engaged in domestic violence or abuse; (4) whether the parent has refused to cooperate with the other parent or the court; (5) whether the parent is paranoid, delusional, or sociopathic; (6) whether the parent believes abuse has occurred; (7) whether the parent feels alienated from the legal system; and (8) whether the parent has a financial reason to stay in the area. The only factor that could arguably implicate race or national origin is whether the parent has strong ties to another country. Yet, even this factor does not necessarily hinge on ethnicity and could apply to a range of circumstances. The factor is relevant merely to determine whether the parent in question could easily relocate.

¶34 Brajesh next asserts the trial court interfered with his “fundamental right to travel abroad with his child.” Pet. for Review at 2. There is no such fundamental right. The “fundamental” right to travel extends only to interstate travel. The United States Supreme Court has explicitly stated foreign travel can be constitutionally limited. Califano v. Aznavorian, 439 U.S. 170, 176, 99 S. Ct. 471, 58 L. Ed. 2d 435 (1978).

¶35 Lastly, Brajesh argues the restrictions interfere with his “fundamental constitutional right to parent his children without state interference” because he is unable to “take his children to India so they can learn about their Indian heritage.” Suppl. Br. of Pet’r at 6. For support, he relies on In re Custody of Smith, 137 Wn.2d 1, 18, 969 P.2d 21 (1998), in which we held state interference with the right *42to rear one’s child requires proof of “some harm [that] threatens the child’s welfare.” As discussed at length above, the trial court explicitly identified the harm involved in this case on the second remand: Brajesh’s credible threats to abscond with the children and his pattern of abusive behavior.

¶36 Furthermore, Smith involved the due process right of a parent against a court’s award of visitation to a nonparent. Id. We have long recognized a parent’s right to raise his or her children may be limited in dissolution proceedings because the competing fundamental rights of both parents and the best interests of the child must also be considered. In re Marriage of King, 162 Wn.2d 378, 388, 174 P.3d 659 (2007) (“[F]undamental constitutional rights are not implicated in a dissolution proceeding.”); Momb v. Ragone, 132 Wn. App. 70, 77, 130 P.3d 406 (2006) (“[N]o case has applied a strict scrutiny standard when weighing the interests of two parents.”). As the Court of Appeals aptly stated below, a parenting plan that “complies with the statutory requirements to promote the best interests of the children” does not violate a parent’s constitutional rights. Katare I, 125 Wn. App. at 823. Because the restrictions imposed by the trial court ultimately complied with RCW 26.09.191(3) and served the best interests of the children, and because the trial court had to balance the constitutional rights of both parents, Brajesh’s constitutional rights as a parent were not violated.

5. Attorney Fees

¶37 Lynette requests attorney fees under In re Marriage of Greenlee, 65 Wn. App. 703, 829 P.2d 1120 (1992), due to Brajesh’s intransigence. “Awards of attorney fees based upon the intransigence of one party have been granted when the party engaged in ‘foot-dragging’ and ‘obstruction’ ... or simply when one party made the trial unduly difficult and increased legal costs by his or her actions.” Id. at 708 (citation omitted) (quoting Eide v. Eide, *431 Wn. App. 440, 445, 462 P.2d 562 (1969)). Lynette argues fees are justified because this is the third appeal in this case and Brajesh has raised variations of the same arguments on every appeal. While we in no way condone Brajesh’s obstinacy, Lynette has not shown his conduct crossed the line to intransigence. We decline to award fees.

Conclusion

¶38 We uphold the travel restrictions imposed by the trial court as they are supported by substantial evidence. Furthermore, the trial court did not abuse its discretion by admitting Berry’s expert testimony regarding risk factors for child abduction. We therefore affirm the Court of Appeals insofar as it upheld the trial court’s parenting plan.

Owens, Fairhurst, and Stephens, JJ., and Alexander, J. Pro Tem., concur.

India is not a signatory to the Hague Convention on the Civil Aspects of International Parental Abduction — a multilateral treaty that provides for summary proceedings in cases of international child abduction. International Parental Child Abduction India, U.S. Dep’t of State (Dec. 2011), http://travel.state.gov/ abduction/country/country_4441.html (last visited July 27, 2012).

RCW 26.09.191(3) provides, in pertinent part:

A parent’s involvement or conduct may have an adverse effect on the child’s best interests, and the court may preclude or limit any provisions of the parenting plan, if any of the following factors exist:

(g) Such other factors or conduct as the court expressly finds adverse to the best interests of the child.

Frye v. United States, 54 App. D.C. 46, 293 F. 1013 (1923).

The risk factors included, inter alia, (1) whether the parent has threatened to abduct or has abducted previously; (2) whether the parent has engaged in planning activities that could facilitate removal of the child from the jurisdiction; (3) whether the parent has engaged in domestic violence or abuse; (4) whether the parent has refused to cooperate with the other parent or the court; (5) whether the parent has strong familial, financial, or cultural ties to another country that is not a party to or compliant with the Hague Convention; (6) whether the parent lacks strong ties to the United States; (7) whether the parent is paranoid delusional or sociopathic; (8) whether the parent believes abuse has occurred; and (9) whether the parent feels alienated from the legal system. See Janet R. Johnston et al., Developing Profiles of Risk for Parental Abduction of Children from a Comparison of Families Victimized by Abduction with Families Litigating Custody, 17 Behav. Sci. & L. 305 (1999); see also Unif. Child Abduction Prevention Act § 7, 9 pt. 1A U.L.A. 50 (Supp. 2011).

The trial court found, in part, “[T]he father threatened to take the children to India without the mother];] . . . [T]he mother’s testimony that the father made threats was credible, when viewed in conjunction with the testimony of others!;] . . . The father sought information for the children in discovery, . . . which would assist in removing the children from the country];] ... The children were too young to seek help if the father improperly retained them in India];] . . . The father’s emails demonstrate extreme anger, abuse, unreasonableness, and poor judgment];] ... The father demonstrated his willingness to punish the children in response to the parenting plan];] . . . India is not a signator to the Hague Convention];] [and] [T]here is no guarantee of enforcing a U.S. parenting order in India.” Clerk’s Papers at 152-57.

The dissent analogizes this case to Abouzahr v. Matera-Abouzahr, 361 N.J. Super. 135, 824 A.2d 268 (2003), to support its position that travel restrictions were improper. The two cases are factually different, however. In Abouzahr, the mother repeatedly stated she trusted her ex-husband: “[The mother] said she did not believe [the father] would retain [their child] in Lebanon beyond the agreed time.” Id. at 140. The mother became concerned only when she learned that Lebanon is not a signatory to the Hague Convention and that the status of *37Lebanese law on child abduction was unfavorable. Id. at 143. Unlike Brajesh, the father in Abouzahr made no specific threats to abduct his child. Instead, the trial court “found credible [the father’s] testimony that he had no intention of abducting [his daughter] or refusing to return her.” Id. at 149. The court declined to impose travel restrictions while noting the difficulty of retrieving an abducted child from a nonsignatory country “is a major factor for the court to weigh ... [b]ut it is not the only factor.” Id. at 156.

The dissent concludes the foreign travel restrictions are not supported by substantial evidence by deconstructing each piece of evidence. Certainly no one piece of evidence alone supports the restrictions. But in the aggregate, and combined with corroborated evidence of Brajesh’s repeated threats to abduct the children, it was not an abuse of discretion for the trial court to conclude Brajesh posed a risk of abduction.

The dissent asserts that the risk factor evidence was improper “because there was no individualization of the risk factors to Mr. Katare.” Dissent at 66. But the factors considered by the trial court were individualized to Mr. Katare. The trial court acknowledged that several risk factors applied specifically to Brajesh but others did not. CP at 156. The dissent fails to offer any insight into what a “proper” method of individualization would look like. Furthermore, even if some of the evidence presented by Mr. Berry was inadmissible — as the dissent posits — in a bench trial, the court is presumed to disregard improper evidence when making its findings. See State v. Miles, 77 Wn.2d 593, 601, 464 P.2d 723 (1970) (noting that in a bench trial there is “a presumption on appeal that the trial judge, knowing the applicable rules of evidence, will not consider matters which are inadmissible when making his findings”).

Concurrence Opinion

Chambers, J.

¶39 (concurring) — I substantially concur with the majority. However, I respectfully disagree with its conclusion that Brajesh Katare’s conduct has not crossed the line to intransigence. Because I conclude it has crossed that line, I would grant Lynette Katare’s request for attorney fees. See In re Marriage of Greenlee, 65 Wn. App. 703, 708, 829 P.2d 1120 (1992) (citing Eide v. Eide, 1 Wn. App. 440, 445, 462 P.2d 562 (1969)).

¶40 Like something out of a Charles Dickens novel, this case has dragged on for years and years. We are currently hearing the third appeal. In the first appeal, the Court of Appeals noted that the trial court’s travel restrictions were supported by the evidence. In re Marriage of Katare, 125 Wn. App. 813, 830-31, 105 P.3d 44 (2004). It remanded merely for clarification in light of an ambiguity in the trial court’s findings. Id. at 831. Since that time, petitioner Brajesh Katare has repeatedly reasserted arguments that had been rejected and has escalated the costs by raising *44new and increasingly extreme arguments. While the trial court found the father’s conduct did not support an award of attorney fees two years ago, the court predicted that it “could support a finding of intransigence in the future.” Clerk’s Papers (CP) at 181. In my view that time has come.

¶41 I certainly agree with many of the sentiments expressed in the chief justice’s dissent. Our courts should not admit evidence based on racial profiling, and we judges absolutely should not make our decisions based on racial animus. “[T]heories and arguments based upon racial, ethnic and most other stereotypes are antithetical to and impermissible in a fair and impartial trial.” State v. Dhaliwal, 150 Wn.2d 559, 583, 79 P.3d 432 (2003) (Chambers, J., concurring); see also State v. Monday, 171 Wn.2d 667, 680, 257 P.3d 551 (2011) (reversing conviction based on prosecutor’s racially charged misconduct); cf. State v. Ladson, 138 Wn.2d 343, 351, 979 P.2d 833 (1999) (excluding evidence seized on pretextual exercises of authority). Some of the expert testimony submitted in this case does not meet these standards. Judges should be quick to sustain objections to testimony that cast cultural and ethnic aspersions, if not take bolder steps. But I am in no way persuaded that Judge Mary E. Roberts based her decision on racially charged factors. Judge Roberts listened to the testimony and concluded that “[t]he risk of abduction by the father and the best interests of the children justify limitations” on foreign travel. CP at 153, 156. Based on that evidence, the judge took steps to prevent Brajesh Katare from taking the children out of the country without their mother’s consent. It is not our role to reweigh the evidence, and I cannot say Judge Roberts abused her discretion.

Dissenting Opinion

Madsen, C.J.

¶42 (dissenting) — The majority sets an unfortunate precedent by permitting improper profile evidence to be admitted on the question of whether a parent may travel with his children to his country of origin without *45an individualized showing that the evidence applies to him. The majority also concludes that sufficient evidence supports the travel restrictions imposed in the parenting plan in this case. I strongly disagree. Based on the record, I believe Mr. Brajesh Katare has been denied important opportunities to share his family and his culture with his children because of his Indian roots.

Procedural History

¶43 This case went through multiple appeals and remands. On Mr. Katare’s third appeal, the Court of Appeals affirmed the trial court’s third try at restricting Mr. Katare’s international travel with his children.

¶44 I find both the reasoning of the trial court and the procedure followed by the Court of Appeals very disturbing. The first obvious problem is that the trial court’s decision to restrict Mr. Katare’s travel with his children started at the wrong end; it started by focusing on the consequences of an abduction and not with the primary question of whether Mr. Katare was likely to abduct his children. At least by the time the trial court found for the second time that Mr. Katare did not pose a serious risk of abduction, the trial court should have concluded that travel restrictions were not necessary. Instead, in each of the three hearings on the matter, the trial court continued to focus on whether adequate remedies existed should Mr. Katare abduct his children rather than the threshold question.

¶45 Similarly, after the trial court’s second finding that Mr. Katare did not pose a serious threat of abducting his children, the Court of Appeals should have reversed the trial court and directed that the travel restriction be removed from the parenting plan because imposing restrictions on a parent who does not present a likely threat of abduction is an abuse of discretion even if obtaining the return of a child from the parent’s country of origin may pose a challenge (also highly debatable here).

*46¶46 In addition to the central question of whether Mr. Katare ever posed a serious risk of abducting the children is the matter of the profile evidence admitted at the third hearing. The Court of Appeals determined that ultimately the trial court did not rely on this profile evidence, and thus its admission was harmless. But the trial court’s written decision shows that it did rely on the improper profile evidence.

¶47 Mr. Katare challenges the trial court’s consideration of the profile evidence as unconstitutional racial profiling. Unlike the Court of Appeals, which concluded this evidence was inadmissible, the majority says it is admissible. The majority fails to properly define the contours of admissibility, however, leaving in place far too broad a rule. The majority also concludes that even to the extent that the profiling evidence might involve racial profiling, sufficient risk factors not concerning race or national origin are established that justify the travel restrictions in this case, and accordingly affirms the Court of Appeals. A close examination of the factual findings of the trial court shows that this is not true.

¶48 In the end, the profile evidence has no place in this litigation because it was never individualized to Mr. Katare himself. The travel restrictions should be stricken because the findings are inadequate to support the conclusion that he poses a serious risk of abduction. Finally, although the trial court entered findings that the law of India does not provide an expeditious, available legal remedy for children improperly abducted under our nation’s laws, the trial court’s findings on this point are contradicted by the very sources on which the court relied. For these reasons, I respectfully dissent.

Analysis

¶49 I agree with the Court of Appeals determination in In re Marriage of Katare, 125 Wn. App. 813, 105 P.3d 44 *47(2004) (.Katare I) that under RCW 26.09.191 a trial court may impose travel restrictions as a component of a parenting plan if the court makes explicit findings that a parent’s conduct justifies the imposition of the restrictions and that the restrictions are “reasonably calculated to address the identified harm.” Id. at 826. The abuse of discretion standard applies, and a trial court’s parenting plan, or a travel restriction included therein, will not be upheld if it is manifestly unreasonable or based on untenable grounds or reasoning. In re Marriage of Littlefield, 133 Wn.2d 39, 46-47, 940 P.2d 1362 (1997).

¶50 The starting point for inquiring into the propriety of the travel restrictions is RCW 26.09.191, which lists factors that may justify limitations on parenting plans:

(3) A parent’s involvement or conduct may have an adverse effect on the child’s best interests, and the court may preclude or limit any provisions of the parenting plan, if any of the following factors exist:

(a) A parent’s neglect or substantial nonperformance of parenting functions;

(b) A long-term emotional or physical impairment which interferes with the parent’s performance of parenting functions as defined in RCW 26.09.004;

(c) A long-term impairment resulting from drug, alcohol, or other substance abuse that interferes with the performance of parenting functions;

(d) The absence or substantial impairment of emotional ties between the parent and the child;

(e) The abusive use of conflict by the parent which creates the danger of serious damage to the child’s psychological development;

(f) A parent has withheld from the other parent access to the child for a protracted period without good cause; or

(g) Such other factors or conduct as the court expressly finds adverse to the best interests of the child.

¶51 The only relevant subsection is (3)(g). Both RCW 26.09.191(3)’s first sentence and RCW 26.09.191(3)(g) ex*48pressly provide that the best interests of the child control whether restrictions may be placed on a parent’s residential time with his children. Importantly, the statute also provides that the particular factor or condition that justifies the restriction must be adverse to the children’s interests. And when a limitation is placed in a parenting plan, the trial court must find a nexus between the parental conduct that is found to support the limitation and an actual or likely adverse impact of the conduct on the children that justifies the restriction. In re Marriage of Watson, 132 Wn. App. 222, 233-34, 130 P.3d 915 (1996).

¶52 Specific to the circumstances here, if the ground for restrictions relates to possible abduction of the child, then, as a Wisconsin court has stated, the best interests of the child standard, because of its breadth, “permits a full consideration of concerns both about a parent’s intention in abducting a child and about the lack of a remedy should that occur.” Long v. Ardestani, 2001 WI App 46, 241 Wis. 2d 498, 528, 624 N.W.2d 405.

¶53 In light of these governing principles, and following a close examination of the findings and evidence regarding the likelihood of abduction, the only possible conclusion in this case is that the trial court’s determination that Mr. Katare poses a serious threat of abduction constitutes an abuse of discretion. In addition, the findings about the adequacy of remedies under the law of India also constitute an abuse of discretion. Finally, admission of the profile evidence in this case constitutes an abuse of discretion and its admission was not harmless because, contrary to the majority, the trial court’s decision is not justified by other findings.

Likelihood of Abduction

¶54 It is important to bear in mind throughout this review that when the original findings were entered in this case the trial court twice expressly found they were insufficient to establish that Mr. Katare posed any serious risk of abducting his children.

*49¶55 Turning to the third hearing, the findings that purportedly justify the imposition of a travel restriction are divided into two categories in the trial court’s final order: the facts that were “brought forth during the June 2003 dissolution trial” and “additional findings based on the evidence presented on remand” in 2009. Clerk’s Papers (CP) at 153-54. Turning to the first of these, the trial court found that Mr. Katare maintains ties to India because he “was born and raised in India,” he has family that lives in India, and “[h]e is now engaged to marry an Indian woman who lives and works in the Seattle area and has applied for a green card.” CP at 153.

¶56 Reliance on a person’s place of birth to support a finding that the parent is likely to abduct his child is improper. A person’s place of birth and family ties do not show whether a parent is likely to abduct his child. Some individuals born in this country and having ties only to this country abduct their children. Conversely, many individuals who have significant ties to a foreign country have no intention of absconding with their children to that country.

¶57 Where a person was born, where their family resides, and who they have chosen to marry are factors that should not be considered unless there is an explicit showing that these factors represent harm to “the children’s physical, mental, or emotional health.” See Watson, 132 Wn. App. at 233-34 (nexus requirement); In re Marriage of Wicklund, 84 Wn. App. 763, 770, 771-72, 932 P.2d 652 (1996) (same; “parental conduct may only be restricted if the conduct ‘would endanger the child’s physical, mental, or emotional health’ ” (internal quotation marks omitted) (quoting In re Marriage of Cabalquinto, 43 Wn. App. 518, 519, 718 P.2d 7 (1986))). Absent a showing of this necessary nexus between factors said to support a restriction in a parenting plan and the children’s best interests, such findings are irrelevant. There is no such showing here, and these first findings are irrelevant to a determination of whether Mr. Katare is likely to abduct his children.

*50¶58 To the contrary, evidence presented to the trial court demonstrated that the Katare children have a deep need to understand their Indian family, culture, and heritage during their childhood after age six, as their identities and self-concepts are forming, and visits to India will serve this need. By ignoring such considerations, the trial court failed to properly consider the best interests of the children as required. “Visitation rights must be determined with reference to the needs of the child rather than the ... preferences of the parent.” In re Marriage of Cabalquinto, 100 Wn.2d 325, 329, 669 P.2d 886 (1983). Similarly, when considering travel restrictions, a court must have in mind the needs of the child. But here the trial court was so focused on the possibility of abduction that it failed to make the proper assessment of the children’s best interests.

¶59 The trial court also found that “in the months leading up to the mother filing a petition for dissolution ... the father threatened to take the children to India without the mother” and the “mother found an application for an Indian PIO [(person of Indian origin)] card ... on the father’s computer.” CP at 153-54. Mr. Katare did consider relocating his family to India in 2002 in order to take advantage of a one-to-two-year-long job opportunity with his United States employer and, as a part of that attempt to relocate his family, he sought PIO cards for his children.

¶60 Ms. Lynette Katare, however, strenuously objected to the relocation because she felt that she would have no life or opportunities if she were to move to India. Thus began a long, difficult disagreement between Mr. and Ms. Katare in which, it is asserted, Mr. Katare threatened that he and the children would relocate to India while his wife stayed in the United States. However, this dispute, which ensued months before the initiation of divorce proceedings, does not indicate that Mr. Katare is now likely to abduct his children. Indeed, Mr. Katare has complied with all court ordered requirements, and there is simply no ground to assume that he would not do so in the future based on statements made before the dissolution proceedings even commenced.

*51¶61 In fact, Mr. Katare did not move his children to India without their mother. Instead, he proposed a one-to-two-year living arrangement in which Ms. Katare could stay in the United States due to her objections to relocation while he and the children lived in India. Moreover, since their predissolution dispute, Mr. Katare has not made any similar threats or suggestions of relocation. Thus, to the extent that his threat to relocate to India prior to his divorce was considered by the court to be an indicator of future attempts to abscond with his children, Mr. Katare has disproved its prediction through his actions and strict adherence to the trial court’s orders.

¶62 The trial court found that during the course of discovery Mr. Katare sought copies of his children’s visas, passport applications, and immunization records — documents that, the court speculates, might assist him in seeking an Indian PIO for his children. CP at 153-54. However, at the time Mr. Katare made these requests there were no travel restrictions in place. The requested documents could legitimately be requested by Mr. Katare simply for the purposes of traveling with his children to visit their grandparents and extended family in India prior to inclusion of the travel restriction in the parenting plan. Therefore, Mr. Katare’s discovery request is not a tenable basis upon which to impose travel restrictions.

¶63 The trial court also made note of the fact that Mr. Katare has the “means and potential to relocate to India for employment.” CP at 154. This finding, however, sheds no light on whether Mr. Katare will abduct his children. Unless there is evidence that Mr. Katare formed an intention to abduct his children and relocate to India, it makes no difference whether it would be financially possible for him to do so, i.e., his ability to do so is irrelevant to the determinative factor of whether he has such an intention in the first place.

¶64 The trial court also found, at the time of the original hearing, that because the Katare children were too young to *52seek help should abduction take place, the consequences of abduction to India are grave and irreversible. Whatever relevance that finding may have had in 2003, by the time of the third hearing in 2009 the Katare children were old enough that they had the ability to use a telephone or access the Internet to let authorities or their mother know if they have been abducted. The trial court’s finding at the latter time that they were still too young is completely undercut by its additional finding that the court did not “place weight on the mother’s attempts to paint her children, who are in gifted programs at school, as incapable of making phone calls or dealing with money.” CP at 155. The court continued, “Her portrayal of the vulnerability of the children was unconvincing to the court.” Id.

¶65 The next set of findings relating to likelihood of abduction are based on the court’s receipt of additional evidence presented on the second remand (the third hearing). CP at 154. They are as follows: a finding that “emails [sent] between the parties after the first trial” indicate that “the father ... harbors resentment against the mother” and show “anger, abuse, unreasonableness, and poor judgment”; a finding that the facts demonstrate Mr. Katare’s willingness to punish his children, with one e-mail given as an example; a finding that Mr. Katare expressed his contempt for the legal system in his correspondence; and a finding that Mr. Katare spent significant time in India since 2003. CP at 154-55.

¶66 The trial court’s finding that Mr. Katare harbors resentment toward Ms. Katare that could manifest itself in abduction is pure speculation. The trial court’s findings establish no basis for concluding that because Mr. Katare resents Ms. Katare, he is likely to abduct the children. The court also found that Mr. Katare addressed Ms. Katare in a condescending and humiliating manner, and doing so while the court is involved shows heightened risk to the children. Again, missing from this finding is any relation to the likelihood of abduction. The e-mails Mr. Katare sent were *53sent to Ms. Katare, and not the children, and we are not concerned in this case with how Mr. Katare deals with his former wife except insofar as how it relates to whether the bests interests of the children are served or disserved by the travel restrictions. RCW 26.09.191(3)(g); Watson, 132 Wn. App. at 233-34. The trial court may be right that Mr. Katare did not behave in a civil manner at all times, but there must be evidence and findings tying his behavior to a risk of abduction, and this tie is absent.

¶67 The trial court’s findings that Mr. Katare is willing to punish his children is evidenced by one e-mail sent from Mr. Katare to Ms. Katare on November 1, 2005, which the trial court included to support its finding. The e-mail reads as follows:

Convey my love and wishes to A and R as today is Diwali. Tell them I love them and they will have their diwali gifts whenever they visit their daddy’s home. They are stored in their play room. Tell them that I will explain what diwali and its significance is [sic] when they grow up.

CP at 154-55. Mr. Katare’s e-mail expresses a desire to be with his children and to teach them about their cultural heritage — something that is in the best interests of children raised in multicultural families. Evidently, however, the court interpreted this e-mail as a father’s withholding of gifts that should have been made available to the children immediately, and delaying of an explanation out of spite or manipulative design. Apparently, Ms. Katare did not view it this way because her e-mail response was to say she would pass the message on to the children, to thank him for sharing, and to wish him “Happy Diwali.” Ex. 37.

¶68 In general, it is not uncommon for a parent of a child to provide a room, furnishings, and toys or other possessions at the parent’s home and not to send everything to other parent’s residence. There is nothing improper about a parent who shares a cultural heritage with the child to want to teach the child about that heritage. More specifi*54cally, here Mr. Katare explains that Diwali is a five-day holiday celebrated in the home with traditional activities and special food and clothing, and a hotel room while exercising visitation rights is not the same. This e-mail may not exemplify what the trial court believes is the best way to share an important holiday with one’s children, but it does not exemplify an improper one and it most certainly does not support the weight given it by the trial court. And again, this finding is not tied to the question whether abduction is likely, as is necessary.

¶69 The trial court’s finding that Mr. Katare harbors contempt for the legal system is based on an e-mail message in which Mr. Katare tells Ms. Katare that if she had not taken the children to Florida, “[I] mean,” he says, “legally abducted” them, then they would not be going through what they were going through. Ex. 15. Mr. Katare obviously disagreed with the decision made by the court to allow Ms. Katare to travel with their children to live in Florida. However, as Mr. Katare has explained in a host of other e-mails, decisions of the court that he might disagree with have not shaken his faith in the system as a whole. Mr. Katare made this particularly clear in an e-mail to Ms. Katare explaining that although he disagreed with particular rulings, “I am sure [the] legal system [will] realize . . . mistakes of reqwarding [sic] your bad behavior. It is only a matter of time.” Id. Further, in a number of e-mails, Mr. Katare has defended the legal system, saying that it is not to be “ma[d]e a joke of” and pledging that “[he] will not violate any court order ever.” Id.

¶70 Much more importantly, Mr. Katare’s actual conduct has reflected the commitment to the law that he has expressed in his correspondence with Ms. Katare. He has, as promised, never violated any order issued by the court during the course of his divorce proceedings. In the face of this contravening evidence, the trial court’s finding concerning Mr. Katare’s contempt for the law, which is premised upon an extrapolation from a single, one-line e-mail, does not support the trial court’s restrictions.

*55¶71 Finally, the trial court’s finding that Mr. Katare “has spent significant time in India” since the 2003 proceedings, CP at 155, does not reasonably indicate that Mr. Katare presents any risk of abduction. As acknowledged by the trial court, even while in India Mr. Katare “kept to the visitation schedule with his children.” Id. That Mr. Katare returned to the United States while in India in order to comply with a court-ordered visitation schedule not only weighs against the trial court’s speculative fear that he will permanently abscond to India, but also shows the lengths to which Mr. Katare will go to see his children while respecting the rulings of the court. Once again, any nexus between Mr. Katare’s conduct of spending significant time in India and any likelihood of abduction is utterly lacking.

¶72 As can be seen, the findings of fact that purportedly address the appropriate question of whether or not Mr. Katare is likely to abduct his children are not supported by substantial evidence and therefore do not justify the trial court’s travel restrictions, or they address matters that are irrelevant or have no apparent nexus to the need for the travel restrictions as required under RCW 26.09.191(3). Accordingly, using these factual findings to support the restrictions constitutes an abuse of discretion. See Wicklund, 84 Wn. App. at 770 n.1; Littlefield, 139 Wn.2d at 47.

¶73 Not only do the specific findings in this case not support the determination that Mr. Katare is likely to abduct his children, the same is true when measured against factors that other courts have considered when assessing the likelihood of abduction. In general, if the likelihood of abduction is high, courts generally impose restrictions to prevent abduction, but when abduction is unlikely, then courts decline to impose preventive measures. Compare Soltanieh v. King, 826 P.2d 1076 (Utah 1992) (where evidence showed that noncustodial parent had no respect for United States laws; did not want his daughter raised under United States standards of education, dress, social relations, political philosophy, and religion; and viewed his daughter and her *56mother as his property and believed himself justified in doing anything necessary to remove his daughter to Iran; made threats; and had had no contact with his daughter for many years, the chance of abduction was high and warranted restrictions), with Abouzahr v. Matera-Abouzahr, 361 N.J. Super. 135, 824 A.2d 268 (2003) (where noncustodial Muslim parent from Lebanon came to the United States for superior medical education and training; married an American Catholic; became a United States citizen; permitted his daughter to be raised in a secular household and to attend Catholic CCD (Cofraternity of Christian Doctrine) classes; lived in the United States for 16 years practicing a medical specialty and teaching at medical schools; brought his mother to live with his wife and child; no evidence showed any disrespect of the United States or its culture, values, and laws; nothing indicated he thought his daughter would have greater values, opportunities, or happiness in Lebanon and instead his words and deeds showed the opposite; and he made no effort after the divorce to sneak his daughter out of the country despite opportunities to do so, the record did not support restriction despite the fact that Lebanon was not a signatory to the Hague Convention on the Civil Aspects of International Child Abduction9).

¶74 The cases are extremely fact specific, of course, and therefore no case from any jurisdiction even hints at a rule for determining the likelihood that a parent will abduct a child. Long, 241 Wis. 2d at 526. However, other jurisdictions have suggested the following factors as relevant considerations when imposing travel restrictions: (1) whether a parent has expressed an intention to abduct his or her own children; (2) whether a parent has had the opportunity to abduct his or her children and whether they attempted to take advantage of that opportunity; (3) whether a parent expresses a disregard for the safety of his or her children; (4) whether a parent has cooperated with the trial court’s *57prior orders; (5) and whether a parent shows marked disapproval of or hostility toward United States cultural, education, political system, values, and religions as they apply to the child. Larisa F. v. Michael S., 120 Misc. 2d 907, 466 N.Y.S.2d 899 (N.Y. Fam. Ct. 1983); Long, 241 Wis. 2d 498; Al-Silham v. Al-Silham, No. 94-A-0048, 1995 Ohio App. LEXIS 5159, 1995 WL 803808 (Ohio Ct. App. Nov. 24, 1995 (unpublished)); Al-Zouhayli v. Al-Zouhayli, 486 N.W.2d 10 (Minn. Ct. App. 1992); Abouzahr, 361 N.J. Super. 135.

¶75 Each of these factors weigh against the trial court’s determination following the third hearing that Mr. Katare is likely to abduct his children. As explained, any statements that could possibly be construed as threats of abduction occurred long ago and have not been made in any recent years. Mr. Katare has also had custody of the children on many occasions and has never, despite having the opportunity to do so, attempted to abduct his children or relocate them to India. He has complied with every court order and has not shown any disregard for the safety of his children. He has exhibited no hostility toward the United States or to his children being raised in this country. To the contrary, he obtained his masters degree here, obtained United States citizenship, has lived and worked in the United States for many years, and continues to maintain residence in the United States while being successfully employed at a company where he has advanced his career over many years. His case is far more like Abouzahr, 361 N.J. Super. 135, than Soltanieh, 826 P.2d 1076.10

*58¶76 Because of the lack of factual support for the trial court’s determination that Mr. Katare poses a serious risk of abduction of his children, this court should reverse the Court of Appeals and remand this case to the trial court for deletion of the travel restrictions from the parenting plan. This should end this case.

Available Remedies If Abduction Occurs

¶77 In the interest of a full assessment, I turn now to the question whether there are adequate remedies if Mr. Katare should, in fact, abduct his children and take them to India to permanently relocate.

¶78 At the outset, I note the trial court’s primary focus in these proceedings has been on the difficulty it believed exists in returning the children to the United States if they are abducted. Indeed, the trial court’s focus on the consequences of abduction as opposed to the likelihood of abduction can be identified at every stage of the proceedings. The court’s 2003 and 2005 assessments of likelihood of abduction highlight the inappropriate consideration given to the consequences of abduction instead of the determinative question of whether Mr. Katare is likely to abduct. In 2003, the trial court was

not persuaded, based on all the evidence presented, including that of the expert witnesses who were called to testify, that Mr. *59Katare presents a serious threat of abducting the children. Nonetheless, if I’m wrong on this the consequences are incredibly serious .... I’m going to impose some restrictions . . . designed to address this issue .... [E]verything that has been brought to this Court... I think indicates that[ ] there is not a serious risk of abduction.

Verbatim Report of Proceedings (VRP) (July 7, 2003) at 10; see also CP at 168 (findings and conclusions, paragraph 2.20.2, to the same effect).11

¶79 In 2005, following the second hearing (after the first remand), the court amended paragraph 2.2 of the parenting plan as follows:

“OTHER FACTORS (RCW 26.09.191(3)). Based on the evidence, including the testimony of expert witnesses, the husband appears to present no serious threat of abducting the children. Nonetheless, under the circumstances of this case, given the ages of the children, the parties’ backgrounds, ties to their families and communities, and history of parenting, and the fact that India is not a signator to the Hague Convention on International Child Abduction, the consequences of such an abduction are so irreversible as to warrant limitations on the husband’s residential time with the children. The risk of abduction is a factor justifying limitations under RCW 26.09.191(3)(g).”

In re Marriage of Katare, noted at 140 Wn. App. 1041, 2007 WL 2823311, at *2, 2007 Wash. App. LEXIS 2755, at *9.

*60¶80 As with its earlier findings, the trial court’s evidentiary hearing on the final remand from the Court of Appeals also exhibits the disproportionate consideration of the consequences of abduction. For example, upon Mr. Katare’s motion to exclude the expert testimony of Michael Berry, the trial judge denied the motion specifically to “allow him to talk about the difficulty in retrieving abducted children [from India].” VRP (Jan. 14, 2009) at 5. Mr. Berry’s testimony concerning the difficulty of retrieving children from India constituted a substantial portion of the court’s evidentiary hearing on remand as well as the court’s findings of fact.

¶81 In its findings following the last of the hearings on the matter, the court entered numerous findings on the availability of adequate remedies, including (1) that “Exhibit 11, at 6.11(3) . . . show[s] the legal impediments to obtaining the return of [a] child ... in India”; (2) that “Exhibit 25 at p.113 show[s] the ... impediments to ... the return of an improperly retained child through the court in India”; (3) that “Exhibit 32, p.8, shows that child abduction is not a crime in India”; (4) that “India is not a signator to the Hague Convention”; (5) that “India has its own laws giving it broad authority to rewrite parenting orders”; (6) that “there is no guarantee of enforcing a U.S. parenting order in India”; (7) that “proceedings in India do not include summary proceedings”; (8) that “proceedings [in India] can take from six months to a year”; and (9) that “the custody order of a foreign state is only one of the factors which will be taken into consideration by a court ... in India.” CP at 155-56.

¶82 Questions regarding foreign law are issues of law that are reviewed de novo on appeal and any trial court finding concerning foreign law must be supported by substantial evidence. State v. Rivera, 95 Wn. App. 961, 966, 977 P.2d 1247 (1999); Byrne v. Cooper, 11 Wn. App. 549, 555, 523 P.2d 1216 (1974) (citing State v. Jackovick, 56 Wn.2d 915, 355 P.2d 976) (1960)).

*61¶83 The trial court’s finding that “proceedings in India do not include summary proceedings” cites exhibit 25. CP at 156. Exhibit 25, however, explicitly states that Indian courts may “ ‘exercise summary jurisdiction in the interests of the child.’ ” Ex. 25 (quoting Dhanwanti Joshi v. Madhave Unde (1997) 1 S.C.C. 112, at *11 (India), available at http://judis.nic.in/supremecourt/helddis.aspx). It also explains alternative measures for timely return of children to their home country:

[A] parent from whom a child has been abducted can petition one of the “State” High Courts to issue a writ of habeas corpus against the abductor ordering the production of the minor in court. This instigates a legal mechanism, similar to the [Hague] Convention, for returning an abducted child to his [/her] country of residence . . . . [I]t . . . allows the petitioner to take advantage of the relative speed and superior authority of the High Court [and] [o]nly circumstances, of which the court is satisfied beyond reasonable doubt, indicating that a return order would inflict serious harm on the child, would merit refusal of an order.

Ex. 25. Unfortunately, exhibit 25 also inaccurately states that summary proceedings are not available in India, citing the Supreme Court of India’s decision in Dhanwanti. But Dhanwanti explains that Indian courts must initially determine whether to “conduct (a) a summary inquiry or (b) an elaborate inquiry on the question of custody.” Dhanwanti, 1 S.C.C. 112, at *9.

¶84 The trial court also found that “Exhibit 11, [section] 6.11(4)... shows that India has its own laws giving it broad authority to rewrite parenting orders of other states.” CP at 156. However, section 6.11(4) of exhibit 11 does not include a single reference to the authority of Indian courts to “rewrite parenting orders” and instead outlines the ways in which Indian courts are actually obligated to follow the parenting orders of other states:

Sections 13 and 14 of the Code of Civil Procedure obliges the courts to give conclusive weight to foreign judgments,... on the following conditions:

*62- the originating court had jurisdiction;

- the merits of the case were considered . . . ;

- international law was correctly applied; and

- the order was not contrary to Indian law.

[T]he Supreme Court of India [has also] elucidated in detail . . . the validity and enforcement of foreign court orders sought to be enforced in India.

Ex. 11.

¶85 The trial court found that abduction proceedings in India can take 6 to 12 months. For this finding, the trial court cites only section 6.11(3) of exhibit 11. CP at 156. Exhibit 11, section 6.11(3), however, describes only the experience of the authors of the text without citation to any authority. CP at 156. This anecdotal and unsubstantiated finding is insufficient to justify the trial court’s restrictions. An analysis of recent Indian case law shows that courts of India may return children abducted from the United States in a very short period of time. See V. Ravi Chandron v. Union of India & Ors., Writ Pet. No. 112/2007 (Supreme Court of India Nov. 17, 2009).

¶86 In short, India’s laws provide a summary procedure. Whether to utilize it may involve more discretion than we are accustomed to in Washington, but it was an error on the trial court’s part to say that it does not exist. From this, the conclusion can only be drawn that the trial court’s findings on the law of India are inaccurate.

¶87 Next, the trial court correctly found that India is not a signatory to the Hague Convention on the Civil Aspects of International Child Abductions. CP at 156. However, the mere fact that a country is not a signatory to this agreement is not in and of itself sufficient to justify travel restrictions absent a contemporaneous showing of a parent’s intention to abduct the child and relocate to that country. “[T]he difficulty of obtaining the return of a child in the event of an abduction (because the other country is not a signatory to the Hague Convention or for other reasons) is [only] one factor *63courts have considered in imposing restrictions [but] in no case ... is this the only factor.” Long, 241 Wis. 2d at 527 (citation omitted). Other jurisdictions have specifically rejected such an argument. See, e.g., Al-Zouhayli, 486 N.W.2d 10 (decision whether to order supervised visitation depends on the facts and circumstances in the case and the unwillingness of a noncustodial parent’s country of national origin to enforce a trial court’s order is not controlling).

¶88 Further, because India has established mechanisms outside the Hague Convention that allow for summary proceedings for return of abducted children to their home country, the importance of this finding is lessened. In any event, absent indication of a contemporaneous showing that a parent intends to abduct his children and relocate to another country, the fact that a country is not a signatory to the Hague Convention is simply irrelevant. Here, no such showing has ever been made.

Profiling Evidence

¶89 Lastly, I turn to the issue of whether the profile evidence should have been allowed in this case. The trial court entered a number of findings about “red flags” in relation to risk factors submitted as part of the profile evidence. The court found that Mr. Katare’s behavior in 2002 and his e-mails, his bitterness toward Ms. Katare, and the lack of resolution of difficulties between the parties “show that he meets the criteria for several Profiles and ‘red flags’ which indicate a risk of abduction by the father, which is against the best interests of the children.” CP at 156.

¶90 The profile evidence was presented through the testimony of Mrs. Katare’s expert, Mr. Berry, and several publications admitted into evidence during his testimony. Among other things, the various risk factors, or psychological profiles to which Berry testified, included such things as the existence of strong emotional or cultural ties to the country of origin, friends or family living in that country, a history of instability in marriage, and a lack of strong ties to the child’s home state.

*64¶91 The Court of Appeals decided that the trial court abused its discretion by admitting this profile evidence because there was no foundation for the testimony and Ms. Katare did not establish that it met the Frye standard for admissibility for novel scientific evidence. In re Marriage of Katare, noted at 159 Wn. App. 1017, 2011 WL 61847, at *11, 2011 Wash. App. LEXIS 65, at *31-32 (Jan 10, 2011) (unpublished); Frye v. United States, 54 App. D.C. 46, 293 F. 1013 (1923). The Court of Appeals said the evidence is analogous to profile evidence that is inadmissible in a criminal case because its probative value is outweighed by prejudicial effect. Katare, 2011 WL 61847, at *11, 2011 Wash. App. LEXIS 65, at *31-32. Such evidence identifies a group of people as being more likely to commit a crime and is inadmissible when it is used to show that a person committed a crime because he shares characteristics with known offenders. 2011 WL 61847, at *12, 2011 Wash. App. LEXIS 65, at *35.

¶92 The majority concludes, however, that such evidence is admissible, analogizing to the “risk assessments [used] to predict the future dangerousness of sexually violent predators.” Majority at 39. The analogy is inapt, however.

¶93 Our precedent concerning the use of profile evidence limits its admissibility to instances in which it is used to create an individualized assessment of risk. Here, the trial court’s findings do not constitute an individualized assessment of risk. Rather, as explained, at the time the profile evidence was admitted in the third hearing, the trial court examined Mr. Katare’s behavior in 2002, finding that Mr. Katare met the profile of an abductor based upon “his behavior in 2002 as shown in Exhibits 39 and 40 and his emails in Exhibit 15, his bitterness . . . and the lack of resolution ... between the parties.” CP at 156. However, the trial court had already reviewed each of these pieces of evidence in the 2003 trial, and at that time concluded that *65they did not indicate that Mr. Katare posed any risk of abducting his children.12

¶94 Accordingly, the trial court’s current finding from the 2009 findings and conclusions can be described only as her inappropriate recognition of Mr. Katare’s similarity to “a group more likely to commit [a] crime” and, therefore, is an impermissible use of profile evidence. State v. Braham, 67 Wn. App. 930, 936, 841 P.2d 785 (1992).

¶95 That the court relied on this evidence is apparent as well. First, contrary to the Court of Appeals’ belief that the trial court appropriately based its decision on its finding that Mr. Katare’s testimony and conduct alone justified the travel restrictions, as explained above the trial court instead indicated it relied on all the findings, including the findings based on the profile evidence. Moreover, Mr. Katare’s testimony and conduct do not support the travel restrictions, as also explained. Finally, since the trial court applied the profiles to evidence that had previously been considered and found to be insufficient to show that Mr. Katare presented a risk of abduction, the only conclusion that can be drawn is that the trial court in fact considered the profile evidence and this is what made it possible for the trial court to enter its contrary finding on this factual question in 2009.

¶96 Finally, much of the briefing from the parties and amici in this court concerns the propriety of racial profiling in child abduction cases. The claims of Mr. Katare and amici weighing in on his side are not without merit. Many of the risk factors identified in the profiling evidence that was admitted in this case will invariably exist with parents *66from specific racial or ethnic backgrounds. For example, in any culture where family ties are important, the risk factor based on strong emotional or cultural ties to family living in that the country of origin will exist. Rather than condemn parents for strong cultural ties and familial ties to their countries of origin, we should, and do, generally speaking, celebrate these qualities rather than use them to restrict a parent’s interactions with his children. It is especially important to avoid basing decisions on such factors when the fact is that we are a nation of many racial and ethnic backgrounds, and we have the good fortune to have many children of mixed race and cultural heritage.

¶97 And finally on this subject, although I believe that the profile evidence was improperly considered in this case for the reasons explained, I also believe that courts must absolutely refrain from engaging in any racial or ethnic profiling that involves conclusions based on race or ethnicity or country of origin. I have previously spoken strongly on this point, and I continue to believe there is no place in our judicial system for such discrimination. See State v. Monday, 171 Wn.2d 667, 682-85, 257 P.3d 551 (2011) (Madsen, C.J., concurring).

¶98 In summary, the profile evidence that was in fact relied on by the trial court should not have been admitted in this case because there was no individualization of the risk factors to Mr. Katare. Instead, the trial court superimposed the risk factors over the evidence previously considered and reached a different result based to a significant degree on the profile evidence. Although the Court of Appeals seems to have established a nearly per se rule of inadmissibility, I would leave the door open for risk factor evidence that is individualized to a parent in proceedings involving conditions and restrictions in a parenting plan. In no event should profile evidence that is racial profiling be permitted.

Conclusion

¶99 The fact that the Court of Appeals did not simply reverse the trial court’s imposition of travel restrictions is *67distressing in this case, when it was obvious after the second hearing on the matter that Mr. Katare did not pose a serious risk of abducting his children, as the evidence showed and as the trial court held. I am positive, however, that Mr. Katare finds it more distressing to be condemned as a potential abductor without sufficient evidence and even more distressing that he cannot take his children to India, where, among other things, their paternal grandparents live. Mr. Katare has demonstrated that he will comply with all court orders and this is backed up by the record, which shows that he has never failed to comply with any court imposed conditions in the years since this litigation began. He has been an attentive father who has fully and compliantly implemented his visitation rights, even traveling from India during a temporary assignment there to visit with his children. He is not an “occasional” parent, but a father with a rich Indian heritage and family in India that he seeks to share with his children.

¶100 I would reverse the Court of Appeals and remand this case to the trial court with directions that it remove the restriction in the parenting plan that presently prevents him from international travel with his children. I dissent.

C. Johnson and Wiggins, JJ., concur with Madsen, C.J.

Hague Convention on the Civil Aspects of International Child Abduction, Oct. 25, 1980, T.I.A.S. No. 11670, 1343 U.N.T.S. 89.

The majority believes that I mistakenly compare this case to Abouzahr. The majority says that the cases are factually different because unlike Mr. Katare, the father in Abouzahr made no specific threats to abduct his child. Majority at 36 n.6. I disagree.

The only reference to “threats” in the trial court’s findings in this case is in the following finding of fact:

• In the months leading up to the mother filing a petition for dissolution of their marriage, the father threatened to take the children to India without the mother. Third parties interviewed by the parenting evaluator stated that they heard the father make similar threats.

*58CP at 153. Thus, the findings say only that before the divorce proceedings began, the father “threatened” to “take the children” to India. Notably, at that time he and the mother disagreed about his taking a temporary job assignment in India. In addition, in the findings of fact in both the first and second proceedings, the trial court found that Mr. Katare was unlikely to abduct his children. The record does not support the trial court’s alteration of its earlier findings that he was not likely to abduct his children.

There are no other findings about threats of abduction or that show that Mr. Katare intended to abduct the children. Thus, contrary to the majority, this case is factually similar to Abouzahr.

Moreover, and also contrary to the majority at 38 n.7, there are no findings of “repeated threats” to abduct and none of the court’s findings support its conclusions. This is not a case where the evidence cumulatively supports the trial court’s order although separately the findings do not. This is a case where none of the findings support the trial court’s conclusions.

The court inserted in the parenting plan the following:

2.20.2 Based on the evidence, including the testimony of expert witnesses, the husband appears to present no serious threat of abducting the children. Nonetheless, under the circumstances of this case, given .the ages of the children, the parties’ backgrounds, ties to their families and communities, and history of parenting, the consequences of such an abduction are so irreversible as to warrant limitations on the husband’s residential time with the children, including: location of exercise of residential time, surrender of his passport, notification of any change of his citizenship status, and prohibition of his holding or obtaining certain documents (i.e. passports, birth certificates) for the children. The mother shall retain the children’s passports.

CP at 168.

The majority believes that the trial court properly individualized the risk factor evidence to Mr. Katare and believes I must show what proper individualization would entail. Majority at 40 n.8. The majority misses my point. The trial court applied profile factors to specific evidence that it had already considered and from which it had already concluded that Mr. Katare did not pose a serious risk of abducting his children. Once the profile evidence was admitted (and it took up the vast part of the third hearing), the court reached the opposite conclusion. Under these circumstances, I do not see how the court could possibly have engaged in a proper individualization of the profile evidence to Mr. Katare.