JOHNSON, J.

The petitioner, Amalgamated Transit Union, Local 717 (union), appeals a decision of the public employee labor relations board (PELRB), in which the PELRB refused to implement an arbitrator's award ordering the respondent, the Manchester Transit Authority (MTA), to reinstate two union employees to their positions after testing positive for drug use by vacating the arbitrator's award. We affirm and remand.

The MTA is a public employer of driver-operators, mechanics, maintenance personnel, and other employees. The union is the duly certified bargaining unit for MTA employees. The MTA and the union entered into a collective bargaining agreement (CBA) that included a grievance process concluding with final and binding arbitration.

The arbitrator found the following facts. In addition to providing general passenger service on established bus routes, the MTA provides bus services under a contract with the Manchester School District for students in the Manchester public schools. MTA buses used for school purposes carry signs indicating that the buses are within a drug-free zone. See RSA 193-B:1, :2 (Supp. 1998).

Ted Urban was employed in the summer of 1992 as a mechanic at the MTA garage. His duties included bus repair and maintenance, retrieval of broken buses, and plowing snow with a large dump truck at the garage lot. Urban tested positive for marijuana in a random drug test. He was suspended on June 26, 1995, and was advised that, under federal regulations issued under the Omnibus Transportation Employee Testing Act of 1991, he would be randomly tested six times in the next year and that he would be terminated if he tested positive a second time. On April 3, 1996, Urban again tested positive. He was suspended on April 11, 1996, when the employer learned he tested positive, and on May 22, 1996, was terminated by the MTA.

Dave Conway was hired by the MTA in September 1985 and worked as a bus driver. Except for one minor, non-drug-related incident in 1987, Conway had a clean disciplinary record with the MTA. Conway's name was randomly selected by computer, and he was taken while driving his route on May 26, 1996, and tested for drug use. On June 6, 1996, the MTA suspended him because his test results were positive for canniboids, and ultimately terminated Conway on June 19, 1996. Conway acknowledged that he smoked a marijuana cigarette the weekend before he was tested.

The arbitrator also found that in 1990, the MTA "adopted a policy stating that use of drugs in the workplace was prohibited and that employees who violated the policy were subject to discipline up to termination." Not until June 26, 1996, however, did the MTA adopt a formal "zero tolerance" drug policy requiring discharge on the first incident of an employee testing positive for drugs.

In February and March of 1997, an arbitrator heard the grievances filed in the Urban and Conway discharges. In both cases, the issues involved whether each employee was discharged for just cause. The arbitrator found that "[n]either grievant was tested because of alleged impairment in his job performance, and there is no evidence that either grievant was at any time impaired because of drug use."

The arbitrator reversed the disciplinary terminations in both cases for several reasons. First, the MTA's "zero tolerance" drug policy, adopted on June 26, 1996, was not in effect when either Urban or Conway was discharged. Second, with respect to the Conway matter, termination after testing positive once was inconsistent with the CBA provision that declared "when discipline is to be given, it shall be given in a fair and progressive manner for repeated offenses." Third, the Manchester School District merely prohibits "[a]ny one, including bus drivers and monitors, caught under the influence (even once)" from interacting with students. Fourth, the arbitrator found that neither employee was under the influence while on the job. Fifth, federal transportation regulations "do not prescribe the level of discipline to be imposed when an employee tests positive for drugs on one occasion without any evidence of impairment." Finally, the arbitrator found that the MTA failed to discipline the employees within the time period prescribed in the CBA.

We need not decide whether the MTA's actions in disciplining these employees contravened the express language of the limitations period outlined in the CBA because, even if it did, we hold that strong public policy would prevent enforcement of that CBA provision.

The arbitrator found that there was "no evidence that either grievant was at any time impaired because of drug use," and ordered the MTA to reinstate the employees to their former positions. The MTA filed an unfair labor practice (ULP) complaint with the PELRB based on the arbitrator's award reinstating the two employees, and the union cross-complained that the MTA committed an ULP by rejecting the arbitrator's decision. The PELRB found the union committed an ULP because the arbitrator's award was contrary to public policy, vacated the arbitrator's award, and dismissed the union's ULP claim. The PELRB found that "in 1990, the [MTA] had adopted a policy prohibiting the use of drugs in the workplace and stating that violators were subject to discipline up to termination."

"[A]dministrative agencies are granted only limited and special subject matter jurisdiction. . . ." 4 R. Wiebusch, New Hampshire Practice, Civil Practice and Procedure § 1.03, at 3 (2d ed. 1997). Because administrative agencies act in a quasi-judicial capacity, see Gould v. Director, N.H. Div. of Motor Vehicles, 138 N.H. 343, 347, 639 A.2d 254, 257-58 (1994), agencies inherently have limited jurisdiction to apply strong and dominant public policy as expressed in controlling statutes, regulations, common law, and other applicable authority, to address matters necessary to resolve questions arising within the scope of their jurisdiction. Just as this court "will not enforce a contract or contract term that contravenes public policy," Harper v. Healthsource New Hampshire, 140 N.H. 770, 775, 674 A.2d 962, 965 (1996), agencies may, within the confines outlined above, do the same. Here, the PELRB recognized a well-established public policy in overruling the arbitrator.

Relying on the dominant public policy against employees in safety-sensitive positions testing positive for drugs discussed in Exxon Corp. v. Esso Workers' Union, Inc., 118 F.3d 841, 848 (1st Cir. 1997), the PELRB ruled that the arbitrator's award could not be enforced because it violates clearly defined public policy. On appeal, the union argues that the arbitrator's award should be enforced because termination is inconsistent with the progressive discipline system established in the CBA, and with the arbitrator's finding that neither employee was impaired while working. After extensive discussion and review of applicable federal statutes, the First Circuit identified a dominant public policy against allowing employees who test positive for drug usage to perform safety-sensitive positions. See Exxon Corp., 118 F.3d at 846-48. We agree with the First Circuit's determination.

Having found that a dominant public policy prohibits reinstatement to a safety-sensitive position, we next determine whether the arbitrator's award violates that policy. The union's argument parallels the argument made by the defendant union in Exxon Corp. In Exxon Corp., a truck driver who tested positive for cocaine usage was dismissed under an established CBA provision allowing discharge for that offense. Id. at 843-44. In both the instant case and Exxon Corp., the union argued that the employee should not be disciplined absent evidence of job-related impairment. Although the employee was terminated in Exxon Corp. because there was a specific CBA provision allowing for termination, we agree with the First Circuit's rationale and reject the union's position in the instant case that the employees should be reinstated absent evidence of job-related impairment. As the First Circuit stated:

According to the Union, the positive result of [the employee's] random drug test "merely" indicates the presence of cocaine in his bloodstream; it does not necessarily signify that [the employee] was under the influence . . . either at the time of the test or at the time he [was performing his work].

The Union casts this argument so narrowly that it misses the mark. Relying upon the job-relatedness as the sole determinative factor in permitting employers to discharge employees who test positive for drug use would force employers to wait for some other consequential indication that drugs are affecting work performance. Typically, this other indication will be an accident. The notorious mishap involving the Exxon Valdez, which produced vast environmental devastation, highlights the core problem associated with this "wait-and-see" approach. If we have learned anything from such catastrophes, it is that employers must act affirmatively to avoid drug-related accidents rather than wait passively for such accidents to happen.

Id. at 849 (citations omitted). In the context of this case, implementation of the arbitrator's award would place two public transit employees — one a driver and the other a mechanic — back in safety-sensitive positions after testing positive for drugs. The union's argument that the MTA should reinstate the employees because there was no evidence of job-related impairment is incredible. Strong public policy may override the terms of a CBA, cf. Paperworkers v. Misco, Inc., 484 U.S. 29, 42-43 (1987), and such a policy prevents reinstatement of these two employees to their positions without a further determination that the employees have ceased drug usage. Cf. Exxon Corp., 118 F.3d at 849. The safety of students and other passengers who rely on MTA transportation on a daily basis cannot be held hostage to the employees' assertion of no job-related impairment. Implementation of the arbitrator's award would allow public transit employees testing positive for drugs to remain in their safety-sensitive positions until the employee was actually impaired on the job, foreseeably evidenced by an accident resulting in a catastrophic loss of life. Although strong public policy prohibits reinstatement without evidence of successful treatment for their drug usage, we agree with the arbitrator that neither the CBA nor dominant public policy prescribes any particular remedy for testing positive for drugs.

Absent a clearly defined zero-tolerance drug policy, such as the 1996 "zero-tolerance" drug policy adopted after the discharges, no dominant public policy compels termination after a positive drug test. Cf. Exxon Corp., 118 F.3d at 849. Although we agree with the PELRB's determination that the arbitrator's decision violated strong public policy by ordering reinstatement and would constitute an ULP if enforced, that same policy preventing reinstatement does not automatically require termination. We do not hold that termination is inappropriate; we hold only that automatic termination is inappropriate simply because the arbitrator's award violates strong public policy. Our review of the record does not reveal that the PELRB considered the appropriate remedy in light of the CBA. The CBA is not before us, and the parties do not argue whether the remedy should be determined by the arbitrator or the PELRB. Thus, we express no opinion on the appropriate remedy, or who should determine it. Accordingly, we remand for a determination of an appropriate remedy. See RSA 273-A:6 (Supp. 1998).

We have reviewed the record with respect to the union's remaining arguments and find them to be without merit and warranting no further discussion. See Vogel v. Vogel, 137 N.H. 321, 322, 627 A.2d 595, 596 (1993).

Affirmed and remanded.

All concurred.