Petitioners, the Deep River Citizens Coalition (DRCC), the Deep River Coalition, Inc. (DRCI), and the American Canoe Association, Inc., appeal an order affirming a final agency decision of the Environmental Management Commission (EMC). The order granted summary judgment against them in a suit involving the construction of a dam on the Deep River in Randleman, North Carolina. Petitioners also appeal a supplemental order delineating the scope of review. For the reasons discussed herein, we reverse and remand.
The facts are as follows: Several North Carolina counties formed the Piedmont Triad Regional Water Authority (Water Authority) in 1986 to manage the region’s water supply needs. In 1988, the Water Authority petitioned the EMC to purchase land and divert 28.5 million gallons of water per day (mgd) from the Deep River Basin to the Haw River Basin pursuant to the power of eminent domain. The EMC approved the inter-basin transfers in 1992 and authorized the Water Authority to use eminent domain to purchase the land needed to construct the dam.
In March 1992, petitioners and other individuals challenged the EMC’s decision. The trial court overturned the EMC’s decision on the basis that the EMC had not resolved water-quality problems and because all impacts and reasonable alternatives had not been ana*213lyzed. However, in 1995, this Court reversed the trial court, stating the trial court did not have jurisdiction because petitioners failed to exhaust their administrative remedies before the Office of Administrative Hearings. See Deep River Citizens’ Coalition v. DEHNR, 119 N.C. App. 232, 457 S.E.2d 772 (1995).
The Water Authority sought to reclassify the portion of the Deep River where the reservoir will be built from Class-C waters to WSTV waters so that it could be used as a water supply. The EMC eventually reclassified portions of Deep River to WS-IV and after applicable certifications were completed, the U.S. Army Corps of Engineers issued a permit for the dam project.
Petitioners filed for a contested case hearing before an administrative law judge (ALJ) challenging the water certification. The AU dismissed petitioner DRCC from the case on the basis that it was not a “person” under the North Carolina Administrative Procedure Act. DRCC then filed a petition for judicial review, but it was stayed pending the determination of the merits of the underlying action. The two cases were consolidated.
In the underlying case, the EMC granted summary judgment to respondents on all issues. Petitioners filed a petition for judicial review. The trial court affirmed the EMC’s decision. Petitioners appealed. On 30 May 2001, the trial court filed a supplemental order concerning the scope of its review. Petitioners also timely appealed from the supplemental order.
By their first assignment of error, petitioners argue the trial court erred in failing to review the EMC’s decision de novo. In examining the trial court’s order for an error of law, this Court will: (1) determine whether the trial court exercised the appropriate scope of review and, if appropriate; (2) decide whether the court did so properly. Eury v. N.C. Employment Security Comm., 115 N.C. App. 590, 597, 446 S.E.2d 383, 388 (1994).
The proper standard of review by the trial court depends upon the particular issues presented by the appeal. ACT-UP Triangle v. Commission for Health Services, 345 N.C. 699, 706, 483 S.E.2d 388, 392 (1997); Brooks v. McWhirter Grading Co., Inc., 303 N.C. 573, 580, 281 S.E.2d 24, 28 (1981). If appellant argues the agency’s decision was based on an error of law, then de novo review is required. In re McCrary, 112 N.C. App. 161, 165, 435 S.E.2d 359, 363 (1993) (citations omitted). If appellant questions whether the agency’s decision was *214supported by the evidence or whether it was arbitrary or capricious, then the reviewing court must apply the whole record test.
In the instant case, petitioners challenged whether the EMC’s conclusions were supported by the record and if DENR’s refusal to conduct a public hearing was an abuse of discretion. These issues focus on whether the EMC’s decision was arbitrary and capricious and whether its decision was supported by substantial evidence. Thus, a whole record review was proper.
In Starco, Inc. v. AMG Bonding & Ins. Services, Inc., 124 N.C. App. 332, 477 S.E.2d 211 (1996), this Court held that it is not improper for a trial court to incorporate documents by reference in its order. The trial court here stated that “in reaching its decision reflected in its Order, the Court adopted and used, in addressing each issue raised by Petitioners, the scope of review under G.S. 150B-51 as urged by Respondent and Respondent-intervenors in their Brief[.]” However, in their brief, respondent and respondent-intervenors request both methods of review. For example, the brief states that
To the extent that the chlorophyll a argument claims that DENR and the EMC misinterpreted their own chlorophyll a rule, it is [sic] should be judged under the de novo review standard, as should Petitioners’ other arguments alleging violations of the North Carolina Environmental Policy Act (“NCEPA”) or DENR’s rules governing the 401 Certification. ... To the extent that [the argument] is a claim that the Decision is arbitrary and capricious or not supported by substantial evidence, [a whole record review should be used.]
This is only a recitation of the general rule. The brief does not address which standard is used for consideration of each specific issue, leaving the trial court room to decide “the extent.” The trial court simply stated it used the scope of review set forth in respondent and respondent-intervenors’ brief without deciding “the extent” it was using each standard. The trial court then concluded the EMC did not err in its decision.
In Hedgepeth v. North Carolina Division of Services for the Blind, 142 N.C. App. 338, 543 S.E.2d 169 (2001), this Court held that
the trial court in the case sub judice stated the proper standards of review sought by petitioner. However, it. . . failed to delineate which standard the court utilized in resolving each separate issue *215raised. Furthermore, it is difficult to discern whether the trial court actually conducted both a “whole record” and de novo review .... We are left to question whether [the trial court] referred to only a “whole record” review, de novo review, or both .... Given the nature of the trial court’s order, we find ourselves unable to conduct our necessary threshold review. And . . . “we decline to speculate in that regard.”
Id. at 349, 543 S.E.2d at 176 (citations omitted). Likewise, in the instant case, the trial court’s supplemental order only states that it used the standard of review set out in section 150B-51, which includes both de novo and whole record reviews. It omits whether it specifically used a de novo or whole record test, and to what extent, for the separate issues raised by petitioners.
Accordingly, we reverse the trial court’s order and remand this matter for a new order in accordance with this opinion. We direct the trial court to: (1) advance its own characterization of the issues presented by petitioners; and (2) clearly delineate the standards of review, detailing the standards used to resolve each distinct issue raised.
REVERSED AND REMANDED.
Judges WYNN and HUDSON concur.