Petitioner WaterWatch of Oregon, Inc. seeks judicial review of three separate final orders issued in contested cases by the Water Resources Department (the department) that were decided based on a consolidated record, and which we have consolidated for purposes of argument and opinion. In those three orders, the department granted to respondents the City of Lake Oswego,1 the South Fork Water Board, and the North Clackamas County Water Commission2 (collectively, the municipal parties) extensions of time to perfect water rights under their respective permits for water diversions from the lower 3.1 miles of the Clackamas River. In granting the extensions, the department was required to condition the “undeveloped portion” of the municipal parties’ permits “to maintain * * * the persistence of fish species listed as sensitive, threatened or endangered under state or federal law.” ORS 537.230(2)(c). Petitioner asserts that the department’s conclusion that the fish-persistence requirement has been met by the conditions that the department placed on the municipal parties’ permits is not supported by substantial evidence and is contrary to law. Petitioner also challenges the department’s modification of the administrative law judge’s findings of fact and the department’s procedural handling of evidence submitted by petitioner.
We conclude that the department’s determination that the permits, as conditioned, will maintain the persistence of listed fish species in the affected waterway lacks both substantial evidence and substantial reason. The department based its decision on the distinction between a short-term drop below persistence flows, which will not affect the persistence of listed fish species, and a long-term drop below persistence flows, which will affect the persistence of listed fish species. However, the record lacks *191substantial evidence of what a short-term drop below persistence flows means versus a long-term drop. Additionally, the department failed to adequately explain how its findings support its conclusion that the undeveloped portions of the permits, as conditioned, will maintain the persistence of the listed fish species when, on their face, the conditions fail to ensure that diversion of the undeveloped portions of the permits will not contribute to long-term drops below persistence flows. We reject all of petitioner’s remaining arguments on judicial review, and we reverse and remand all three final orders to the department for further consideration.
I. BACKGROUND
A. Overview
The municipal parties are holders of eight separate water-right permits for municipal use that have points of diversion in the lower 3.1 miles of the Clackamas River (affected reach or lower reach).3 The holder of a permit for municipal water use must complete construction of any works within 20 years of obtaining the permit and put the water right to complete use within the time frame specified in the permit. ORS 537.230(2). A municipal water holder can obtain an extension of time of those deadlines if the department finds that three statutory conditions have been satisfied.4 Id. At issue in these cases is the department’s *192application of the third statutory condition in granting the municipal parties’ requested extensions.
That condition required the department to find that the “undeveloped portion” of the municipal parties’ permits are “conditioned to maintain, in the portions of waterways affected by water use under the permit, the persistence of fish species listed as sensitive, threatened or endangered under state or federal law.” ORS 537.230(2)(c). The statute requires the department to “base its finding on existing data and upon the advice of the State Department of Fish and Wildlife.” Id. By rule, the department’s finding is limited to effects “related to streamflow as a result of use of the undeveloped portion of the permit and further limited to where, as a result of use of the undeveloped portion of the permit, [the Oregon Department of Fish and Wildlife (ODFW)] indicates that streamflow would be a limiting factor for the subject listed fish species.” OAR 690-315-0080(2).
With that background in mind, we turn to the record developed below.
B. Water permits
1. Lake Oswego
The City of Lake Oswego is the holder of two water right permits at issue in these cases. Permit S-32410, which was granted to the city on October 19, 1967, authorizes the city to use up to 50.0 cubic feet per second (cfs) of water from *193the Clackamas River. “Construction of the water development project was to be completed by October 1, 1969, and complete application of water was to be made on or before October 1, 1970.” Permit S-37839, which was granted to the city on June 27, 1975, authorizes the city to use up to 9.0 cfs of water from the Clackamas River. “Construction of the water development project was to be completed by October 1, 1977, and complete application of water was to be made on or before October 19, 1978.”
The city had received multiple prior extensions of times for both permits, with the most recent extensions setting all permit deadlines on October 1, 2000. On July 1, 2003, the city submitted to the department the current requests to extend the deadlines on both permits to October 1, 2040. The undeveloped portion of Permit S-32410 is 25.0 cfs, and Permit S-37839 is 9.0 cfs.
2. North Clackamas County Water Commission
North Clackamas County Water Commission (North Clackamas)5 is the holder of three water-right permits at issue in these cases. Permit S-46120, which was granted to the City of Gladstone on January 18, 1982, and transferred to North Clackamas in 2005, authorizes North Clackamas to use up to 8.0 cfs of water from the Clackamas River. “Construction of the water development project was to be completed by October 1, 1983, and complete application of water was to be made on or before October 1, 1984.” The department granted two prior extensions of time for Permit S-46120, with the most recent extending all permit deadlines to October 1, 1993. On June 22, 2006, North Clackamas submitted an application to extend Permit S-46120 to October 1, 2025. The undeveloped portion of that permit is 2.99 cfs.
Permit S-35297, which was granted to Oak Lodge Water District on August 25,1971, and transferred to North Clackamas in 2004, authorizes North Clackamas to use up to 62.0 cfs of water from the Clackamas River. “Construction *194of the water development project was to be completed by October 1,1973, and complete application of water was to be made on or before October 1,1974.” The department granted multiple prior extensions of time for Permit S-35297, with the most recent extending all permit deadlines to October 1, 2000. In April 2003, Oak Lodge Water District (predecessor to North Clackamas) submitted an application to extend Permit S-35297, which North Clackamas amended in 2005 to request an extension to October 1, 2030. The undeveloped portion of that permit is 29.01 cfs.
Permit S-43170, which was granted to the City of Gladstone on July 25, 1978, and transferred to North Clackamas in 2005, authorizes North Clackamas to use up to 1.73 cfs of water from the Clackamas River. “Construction of the water development project was to be completed by October 1,1980, and complete application of water was to be made on or before October 19,1981.” The department granted multiple prior extensions of time for Permit S-43170, with the most recent extending all permit deadlines to October 1, 2000. In June 2006, North Clackamas submitted an application to extend Permit S-43170 to October 1, 2025. The undeveloped portion of that permit is 1.73 cfs.
3. South Fork Water Board
South Fork Water Board (South Fork)6 is the holder of three permits at issue in these cases. Permit S-3778, which was granted to the City of Oregon City on May 11,1918, and transferred to South Fork in 1983, authorizes South Fork to use up to 20.0 cfs of water from the South Fork Clackamas River, a tributary of the Clackamas River. “Construction of the water development project was to be completed by May 11, 1923, and complete application of water was to be made on or before October 1,1943.” The department granted multiple prior extensions of time for Permit S-3778, with the most recent extending all permit deadlines to October 1, 2000. In December 2003, South Fork submitted its application to extend Permit S-3778 to October 1, 2050. The undeveloped portion of that permit is 15.0 cfs.
*195Permit S-9982, which was granted to the Cities of Oregon City and West Linn on January 19, 1931, and later transferred to South Fork, authorizes South Fork to use up to 30.0 cfs of water from tributaries to the Clackamas River (20.0 cfs from the South Fork Clackamas River and 10.0 cfs from Memaloose Creek). The permit did not contain construction or full application dates and, as a result, the department originally advised South Fork that it did not have to obtain extensions of time for the permit. However, in August 2004, the department advised South Fork that it would need to apply for an extension. In August 2006, South Fork submitted an application to extend Permit S-9982 to October 1, 2038. The undeveloped portion of that permit is 27.0 cfs.
Permit S-22581, which was granted to the South Fork Water Commission (the predecessor to South Fork) on January 22, 1954, authorizes South Fork to use up to 60.0 cfs of water from the Clackamas River. “Construction of the water development project was to be completed by October 1, 1955, and complete application of water was to be made on or before October 19, 1956.” The department granted multiple prior extensions for Permit S-22581, with the most recent extending all permit deadlines to October 1, 1999. In December 2003, South Fork submitted an application to extend deadlines to October 1, 2049. The undeveloped portion of Permit S-22581 is 37.6 cfs.
C. ODFW advice and the department’s proposed final orders
Pursuant to the fish-persistence requirement in ORS 537.230(2)(c), and after receiving supplemental information from the municipal parties, the department forwarded all eight extension applications to ODFW to review the effect of development of the undeveloped portions of the permits on listed fish species in the lower reach of the Clackamas River. In May 2007, ODFW issued letters to the department for each of the municipal parties’ eight applications that took into account the effect of all eight extension requests and contained identical advice.7 ODFW identified *196target streamflows for fish persistence for each season (persistence flows) and advised the department to condition the undeveloped portion of each of the permits to “maintain persistence of listed fish species consistent with the recommended flows.”
From April to June, ODFW identified a persistence flow of 800 cfs, decreasing to 650 cfs in June. It advised that,
“[i]f flows do not meet targets, [the permit holder] should develop a plan to provide for a contingency to reduce its water use. However, according to gaged records this would be an extremely rare event that has not occurred in the gaged record. The severity of the measures taken should be reflective of how much the recommended flows are being missed by and the percentage of water that is withdrawn by the municipality as compared to the overall streamflow level.”
For July and August, ODFW identified a persistence flow of 650 cfs, which, on occasion, would not be met. ODFW advised:
“If flows do not meet targets, [the permit holder] should develop a plan to provide for a contingency to reduce its water use or augment stream flows using releases from Timothy Lake. Following are considerations for the Water Resources Department to consider in developing conditions for this permit and for the municipality to consider in the development of any plan to address short falls in stream flow levels.
“• If targeted flow levels cannot be met, flow releases under agreement from Timothy Lake can be beneficial to stream flows and can offset some of the use' by the municipalities. Consultation with ODFW is recommended to determine the annual priority for shaping the augmentation flows to best support fish persistence. A plan (in consultation with ODFW) should be developed that considers a flow regime that considers and balances flow augmentation to maintain inundation of winter steelhead redds in early July (through approximately July 15) and the maintenance of consistent flows throughout the remainder of the time period to maximize access to rearing habitat and avoid stranding of fish.
*197“• The severity of the measures taken should be reflective of the available summer rearing habitat within the lower 3.1 miles of the Clackamas River where the diversions occur (which represents less than 2% of the total available rearing habitat) and is habitat that may be avoided by salmonids since the highest temperatures in the basin occur in this stream reach. Because the value of this rearing habitat is low relative to the rest of the basin rearing habitat, measures such as flow augmentation using stored water will offset much of the effect of diverting water out of stream. Additionally, flow augmentation would benefit streamflows and rearing habitat from Timothy Lake through the entire stream reach (23.3 miles) down to the lower 3.1 stream miles where water is withdrawn.”
(Emphasis and boldface in original.)
From September to November, ODFW identified a persistence flow of 650 cfs to September 15 and 800 cfs after September 15. ODFW advised:
“If flows do not meet targets after the first Monday in September, [the permit holder] should develop a plan to provide for a plan to augment stream flows and reduce its water use to minimize its impact. Following are considerations for the Water Resources Department to consider in developing conditions for this permit and for the municipality to consider in the development of any plan to address short falls in stream flow levels.
“• If targeted flow levels cannot be met, flow releases under agreement from Timothy Lake can be beneficial to stream flows and can offset some of the use by the municipalities. A main consideration for this time period is to balance flow augmentation to provide for increasing flows that once reached will not be reduced before fall rains arrive and stream flows naturally begin rising. A plan (in consultation with ODFW) should be developed that considers a flow regime that works best for fish spawning in the lower river that provides access to spawning areas and maintains water over those spawning areas until stream flows naturally increase in the fall.
“• The severity of the measures taken should be reflective of how much the recommended flows are being *198missed by and the percentage of water that is withdrawn by the municipality vs. the overall streamflow level.
“• Relative to the summer flow season, the significance of the lower 3.1 miles in terms of habitat in the fall is more significant (especially for Fall Chinook) and is more important in maintaining persistence of listed and sensitive species.”
(Emphasis and boldface in original.)
From December to March, ODFW identified a persistence flow of 800 cfs, and it advised that it did not anticipate flow-related issues during that time of year. However, ODFW recommended that, “if targeted flow levels cannot be met[,] then the severity of the measures taken should be reflective of how much the recommended flows are being missed by and the percentage of water that is withdrawn by the municipality vs. the overall streamflow level.”
After receiving that advice, in November 2007, the department issued proposed final orders (PFOs) that granted the requested extensions of time on all eight of the municipal parties’ permits. Based on ODFW’s advice, the department found that use of the undeveloped portions of the permits would not maintain the persistence of listed fish species in the affected reach and thus imposed conditions on the permits to maintain fish persistence. Those conditions included ODFW’s recommended minimum flow levels by season and set out the following conditions:
“a. Minimum fish flow needs on the Lower Clackamas River as recommended by ODFW are in Table 2, below, and are to be measured at USGS Gage Number 14211010, Clackamas River near Oregon City, Oregon, or its equivalent.
“b. In cooperation with other members of the Clackamas River Water Providers, [the permit holder] must have an annual meeting with ODFW to devise a strategy to maximize fishery benefits that can be derived from the agreement with PGE for the release of stored water from Timothy Lake. This is of particular significance when augmenting stream flow during the period of July 1 through November 30.
*199“c. From the first Monday in September through June 30 the maximum total amount of the undeveloped portion of the [permit] that can legally be diverted shall be reduced in proportion to the amount by which the flows shown in Table 2 are not met based on a seven day rolling average of mean daily flows (measured on the Clackamas River at USGS Gage Number 14211010, Clackamas River near Oregon City, Oregon, or its equivalent), as illustrated in the examples below.”
ODFW concurred in a one-sentence email that the conditions included in the PFOs were consistent with ODFW’s advice to the department.
D. Contested case hearings
Both petitioner and South Fork protested and requested a contested case hearing for all eight extension orders, which were then consolidated by the administrative law judge (ALJ). In March 2010, the ALJ held a three-day contested case hearing that was split into four parts. The first part addressed issues common to all eight applications, including matters related to fish persistence, and the remaining three parts addressed issues individual to each permit holder.
Before holding the hearing, the ALJ addressed matters raised by the parties in their motions for summary determinations. As relevant here, the ALJ concluded that ORS 537.230 does not require the department to consider climate change in setting fish-persistence conditions. In accordance with that determination, at the hearing, the ALJ excluded from evidence the exhibits that petitioner offered related to the effect of climate change on Oregon’s municipal water supplies and fish and wildlife. The ALJ also excluded as irrelevant letters authored by the Oregon Department of Environmental Quality (DEQ) in 2001 and 2003 that recommended that the department deny applications (not at issue in these cases) for water diversions from the Clackamas River because the lower reach was not meeting water quality standards for temperature, which negatively affects fish. During the hearing, petitioner made oral offers of proof with regard to the DEQ letters. After the contested case hearings, petitioner submitted written offers of proof concerning its climate-change exhibits.
*200In August 2010, the ALJ issued three separate proposed orders — one for Lake Oswego’s permits, one for North Clackamas’s permits, and one for South Fork’s permits — that rejected each of the protests and affirmed the department’s PFOs. The three orders contained the same conclusions and recommendations with respect to the fish-persistence requirement in ORS 537.230(2)(c). The ALJ interpreted that statutory requirement as “leaving] the [department with no option but to follow the advice from ODFW.” Based on that view of the statute, the ALJ concluded that he could not “go behind” the advice provided by ODFW to determine whether it was correct or incorrect. Because the ALJ found that the department had obtained and applied ODFW’s advice, with “one exception,” he concluded that, “[a]s [the department] argued, the inquiry about fish persistence of listed fish stops there.”
The one exception the ALJ had to the department’s PFOs was a modification to the yearly meeting condition. The ALJ recommended the following modification:
“First, although [the department] anticipated an informal meeting with no written conclusions, ODFW intended that there be a written agreement from the meetings— something that the State and the municipalities could look at and use as their guideline for that year. After the evidence was presented, [the department] agreed that a written record of the meeting was appropriate.
“Second, although the conditions are written to require a meeting in which ODFW and the municipality agree as to what should be done in a given year, the condition should be clarified to address the situations in which ODFW and the municipality are not able to reach an agreement.
“*** BaSed upon the importance the Legislature has placed on maintaining the persistence of listed fish, the PFO conditions should be clarified to require the municipalities to accede to ODFW’s fish persistence standards if agreement cannot be reached.”
All parties, including the department, filed exceptions to the ALJ’s proposed orders.
*201E. The department’s amended proposed orders and final orders
In January 2011, the department issued amended proposed orders that incorporated much of the ALJ’s proposed orders, but also modified and added to them in significant ways.
The most significant modification made by the department pertained to what it understood its role to be with regard to the fish-persistence requirement in ORS 537.230(2)(c). The department deleted the ALJ’s entire analysis on that issue and summed up its own conclusions as follows:
“ORS 537.230(2)(c) establishes two sources of information upon which [the department] must base its fish persistence finding. Those sources of information (ODFW’s advice and ‘existing data’) will either be consistent or inconsistent. When ODFW’s advice and existing evidence are consistent, [the department] must adopt conditions consistent with that advice and existing data. However, if ODFW’s advice requires restrictions on water use greater than the existing data demonstrates, by a preponderance of the evidence, is necessary to maintain the persistence of listed fish species, [the department] may deviate from ODFW’s advice. In this circumstance, [the department] may adopt conditions that, based on the existing data, are sufficient to maintain the persistence of listed fish species. And in a case where existing data demonstrates, by a preponderance of the evidence, that conditions consistent with ODFW’s advice would be insufficient to maintain the persistence of listed fish species, [the department] must deviate from ODFW’s advice. In this case, [the department] must adopt conditions that will maintain the persistence of listed fish species, as supported by existing data.”
(Emphases in original.)
The department also admitted and considered petitioner’s climate-change exhibits that the ALJ had excluded from evidence. The department interpreted its rule that limits the scope of the department’s findings to streamflow effects, OAR 690-315-0080(2), “to permit data pertaining to the effects of climate change on future streamflow to be *202considered as a part of the baseline river condition.” However, the department agreed that the ALJ properly excluded the DEQ letters from evidence.
Based on the above conclusions, the department adopted all of the ALJ’s findings of fact, with one partial sentence deletion, but also made 35 additional findings of fact.8 Those findings, where relevant to petitioner’s arguments, are discussed below. The department explained that it made the additional findings because it disagreed with the ALJ’s conclusion that the department was required to accept ODFW’s advice, even if the weight of “existing data” merited a different conclusion. The department thus made findings pertaining to the submitted fish-persistence data, including petitioner’s climate-change exhibits, as well as the ODFW advice.
The department then concluded that, in these cases, ODFW’s advice was based on existing data, the department’s conditions were based on that advice, and ODFW’s concurrence with the department’s conditions was substantial evidence that the department’s “finding and conditions will result in the maintenance of the persistence of listed fish species.” The department also concluded that any party could submit additional existing data to demonstrate that the weight of evidence requires different conditions, but that petitioner had not met that burden here.
With regard to the annual meeting condition that the ALJ had modified, the department agreed that the condition as stated in the PFOs “failed to fully incorporate the ODFW Advice.” The final conditions placed on each of the municipal parties’ permits included a modified meeting condition, as follows:
“a. Minimum fish flow needs on the Lower Clackamas River as recommended by ODFW are in Table 1, below, and are to be measured at USGS Gage Number 14211010, Clackamas River near Oregon City, Oregon, or its equivalent.
*203“b. In cooperation with the holders of Permits S-46120, S-35297, S-43170, S-22581, S-3778, S-9982, S-32410 and S-37839, the permittee must have an annual meeting with ODFW to devise a strategy to maximize fishery benefits that can be derived from the agreement with PGE for the release of stored water from Timothy Lake. It is [the department’s] intent that ODFW and the permittees shall reach agreement on the strategy. However, if after making a good faith effort ODFW and the permittees are unable to reach agreement on a strategy, ODFW shall devise the strategy. In either case, the strategy shall be documented in writing and the permittees shall comply with the strategy. The annual meeting and resulting strategy may cover issues other than Timothy Lake releases that are relevant to both use under Permits S-46120, S-35297, S-43170, S-22581, S-3778, S-9982, S-32410 and S-37839, and to listed fish species; however, the strategy may include actions pertaining to such issues only upon mutual agreement by ODFW and the permittees.
“c. From the first Monday in September through June 30 the maximum total amount of the undeveloped portion of the [permit] that can legally be diverted shall be reduced in proportion to the amount by which the flows shown in Table 1 are not met based on a seven day rolling average of mean daily flows (measured on the Clackamas River at USGS Gage Number 14211010, Clackamas River near Oregon City, Oregon, or its equivalent), as illustrated in the examples below.”
Petitioner and the municipal parties filed exceptions to those amended proposed orders. The department then issued, in April 2011, the three final orders that are the subject of our judicial review here. In the final orders, the department adopted the amended proposed orders with three corrections that did not change the substance of the amended proposed orders.9 Petitioner timely filed its *204petitions for judicial review of those three final orders. As noted, we have consolidated petitioners’ three petitions for the purpose of argument and opinion.
On judicial review, petitioner raises five assignments of error: (1) the department’s findings regarding fish persistence are not supported by substantial evidence or substantial reason; (2) the department’s application of the statutory fish-persistence requirement in the final orders is contrary to law; (3) the department unlawfully modified the AL J’s findings of fact; (4) the AL J failed to follow prescribed administrative procedures, which unlawfully impaired the fairness of the proceedings; and (5) the department unlawfully considered petitioner’s excluded climate-change exhibits without first reopening the contested case hearing before the ALJ and unlawfully excluded the DEQ letters.
II. CONSTRUCTION OF ORS 537.230(2)(c)
We first address petitioner’s second assignment of error, which requires us to construe the fish-persistence requirement in ORS 537.230(2)(c). As noted, that provision provides:
“[T]he department may order and allow an extension of time to complete construction or to perfect a water right beyond the time specified in the permit under the following conditions:
“(c) *** the department finds that the undeveloped portion of the permit is conditioned to maintain, in the portions of waterways affected by water use under the permit, the persistence of fish species listed as sensitive, threatened or endangered under state or federal law. The department shall base its finding on existing data and upon the advice of the State Department of Fish and Wildlife. * *
Petitioner argues that the department’s interpretation and application of ORS 537.230(2)(c) in its orders was contrary to law. Because the disputed statutory phrase is part of a regulatory scheme administered by the department, our standard of review depends on whether the disputed phrase is an exact term, an inexact term, or a delegative term. Springfield Education Assn. v. School Dist., 290 *205Or 217, 223, 621 P2d 547 (1980). The Supreme Court has summarized the different standards:
“‘Exact terms’ impart relatively precise meanings, and their applicability in a particular case involves only agency factfinding. [Springfield Education Assn., 290 Or] at 223-24. This court reviews agency application of ‘exact terms’ for substantial evidence. Id. at 224. ‘Inexact terms’ are less precise. Although they embody a complete expression of legislative meaning, that meaning always may not be obvious. Id. As to ‘inexact terms,’ the task of the agency, and ultimately of the court, is to determine what the legislature intended by using those words. Id. ‘Delegative terms’ express incomplete legislative meaning that the agency is authorized to complete. Id. at 228. As to ‘delegative terms,’ the agency’s task is to complete the general legislative policy decision. Id. This court reviews the agency decision concerning a ‘delegative term’ to determine whether it is within the range of discretion allowed by the more general policy of the statute. Id. at 229.
Coast Security Mortgage Corp. v. Real Estate Agency, 331 Or 348, 353-54, 15 P3d 29 (2000).
Petitioner and the department agree that the phrase — “maintain * * * the persistence of [listed] fish species”— is an inexact term under Springfield. The municipal parties, however, argue that it is a delegative term. We agree with petitioner and the department that the phrase is an inexact term. It is a phrase that expresses a complete legislative policy to ensure that further development of municipal permits will maintain fish persistence, but it is not so precise that it is an “exact term.” We reject the municipal parties’ contention that the phrase must be delegative merely because the legislature did not set forth factors for the department to consider in setting conditions. The absence of factors does not negate that the legislature expressed a complete policy for the department to implement. That the agency must use judgment to determine what conditions it will need to impose on individual applications for extensions of time to effect that complete policy statement is what makes the term inexact. Thus, we must determine whether the department’s interpretation and application of ORS 537.230(2)(c), as embodied in its final orders, is consistent with the *206legislature’s intent in using the words that it did. Coast Security Mortgage Corp., 331 Or at 354.
In discussing the construction of ORS 537.230(2)(c), the parties focus solely on the phrase “maintain * * * the persistence of [listed] fish species.” However, in construing that phrase, we must also take into account the context of the surrounding language, which provides that the department may allow an extension of time if, among other things, “the department finds that the undeveloped portion of the permit is conditioned to maintain, in the portions of waterways affected by water use under the permit, the persistence of [listed] fish species.” What the two other major phrases of the statute mean has already been addressed. In a prior case, we concluded that “[t]he text, context, and legislative history of ORS 537.230(2) indicate that the ‘undeveloped portion of the permit’ is to be measured by reference to the maximum rate of water applied to beneficial use before the expiration of the development deadline in the permit or last-issued extension.”10 WaterWatch of Oregon, Inc. v. Water Resources Dept., 259 Or App 717, 742, 316 P3d 330 (2013), rev allowed, 355 Or 317 (2014). And, the department lias defined by rule that the “portions of waterways affected by water use under the permit” means “those portions of the drainage basin at or below the point of diversion for a surface water permit.” OAR 690-315-0010(6)(f). The department, however, has not promulgated a rule to define or clarify the meaning of the phrase, “is conditioned to maintain * * * the persistence of [listed] fish species.” The construction of the phrase as used in ORS 537.230 (2)(c) is a matter of first impression.
We begin with the plain meaning of the phrase. “Conditioned,” when used as a verb, means “to invest with, limit by, or subject to conditions : burden with a condition : make conditional.” Webster’s Third New Int’l Dictionary 473 (unabridged ed 2002) (boldface in original). “Condition,” as a noun, in turn means “la: something established or agreed upon as a requisite to the doing or taking effect of something else : STIPULATION *** 2 : something that exists *207as an occasion of something else : *** PREREQUISITE.” Id. (capitalization and boldface in original). “Maintain” means “to keep in a state of repair, efficiency, or validity : preserve from failure or decline.” Id. at 1362 (boldface in original). The applicable definition of “persistence” is “the quality or state of being persistent: as * * * continued existence.” Id. at 1686. “Persist,” in turn, means “to continue to exist or endure.” Id. From those definitions, we conclude that the legislature intended that the undeveloped portions of the permits be subject to conditions — that is, fulfillment of the conditions are a prerequisite to diversion of the undeveloped portions — -that preserve from decline the continued existence, or endurance, of listed fish species. The legislative history supports that plain-text understanding.
We conducted a lengthy review of the legislative history of the enactment of ORS 537.230(2)(c) in WaterWatch, 259 Or App at 737-41. In short, the legislature amended ORS 537.230 in 2005 following our decision in WaterWatch v. Water Resources Commission, 193 Or App 87, 88 P3d 327 (2004), vac’d and rem’d, 339 Or 275, 119 P3d 221 (2005), to address concerns that the “decision was contrary to [the department’s] long-standing interpretation and application of the statute * * * and thus raised concerns about the viability of existing municipal permits and the ability of municipalities to plan for future water supply.” WaterWatch, 259 Or App at 738. The original bill did not include the fish-persistence condition found in subsection (2)(c).
“The final condition — what eventually became subsection (2)(c) of ORS 537.230 — was added to the bill by the Senate Committee on Environment and Land Use, see Senate Amendments to A-Engrossed House Bill, HB 3038, June 13, 2005, § 1, and represented a compromise between WaterWatch, the department, and representatives of municipal water suppliers, including the League of Oregon Cities. Testimony, Senate Committee on Environment and Land Use, HB 3038, June 2, 2005, Ex M (statement of Doug Meyers, WaterWatch of Oregon). ***
“WaterWatch presented testimony that, although ‘less than ideal,’ the amendments ‘at least explicitly require some consideration of environmental impacts before extending *208the time to develop a municipal appropriation permit issued before November 2, 1998/ and require the department ‘to deny extensions of such permits unless the extensions are conditioned to maintain populations of fish listed as sensitive, threatened, or endangered under state or federal law.’ Testimony, Senate Committee on Environment and Land Use, HB 3038, June 2, 2005, Ex M (statement of Doug Meyers). Emphasizing its support for subsection (2)(c), Meyers, on behalf of WaterWatch, explained that it ‘requires the department to consult with [the Oregon Department of Fish and Wildlife] and requires the decision to be based on all existing data which can be brought forward in the existing agency process that allows public participation in extension proceedings.’ Tape Recording, Senate Committee on Environment and Land Use, HB 3038, June 2, 2005, Tape 105, Side B (statement of Doug Meyers).
“Also speaking to the compromise, Adam Sussman for the department testified that subsection (2)(c) was designed to provide a ‘resource protection baseline’ to ‘ensure that the undeveloped portion’ of the permit would maintain fish persistence. Tape Recording, Senate Committee on Environment and Land Use, HB 3038, June 2, 2005, Tape 105, Side B (statement of Adam Sussman).”
Id. at 739, 740 (brackets in original).
In addition, in statements on the Senate and the House floor, Senator Ringo and Representative Jenson specifically addressed the intended meaning of the phrase “maintain * * * the persistence offish species.” On the Senate floor, Senator Ringo stated:
“A particular issue, the bill has the specific words that it says the work group agreed that the definition of, ‘maintaining the persistence of fish/ those words are in the bill ‘maintaining the persistence of fish/ and I just want to state for the record what that means, that that phrase, ‘maintaining the persistence of fish/ uses the definition that is contained in the Oregon Plan and that means it’s a forecast of future population health stated in terms of the probability of extirpation. * * * Again this was an agreement that was reached in the workgroup.”
Tape Recording, Senate Floor Debate, HB 3038, June 15, 2005, Tape 186, Side B (statement of Sen Charlie Ringo). On the House floor, Representative Jenson stated:
*209“[A]s a consequence and as an effort to get WaterWatch to come on board with the bill, the Senate committee added some language, language that you would see in section 1,2c * * * it is acceptable language with one particular notation that we talked to the chair of the senate committee about and that he agreed to and that he read into the record on the floor of the Senate before they passed the bill and that I would like also to read into the record, the understanding that was present when this agreement was, when this issue was agreed to by both the cites, and municipal water uses, and Senate committee, and House Water Committee. And that is in line 5 of the 2nd page there is the use of the word, * * * that language is that the portion of the permit is conditioned to maintain, and the word maintain was of some concern to us, and so everybody agreed that as far as this wording is concerned the understanding is it’s to maintain a future population health stated in terms of the probability of extirpation relating to viability.”
Tape Recording, House Floor Debate, HB 3038, June 17, 2005, Tape 136, Side B to Tape 135, Side B (statement of Rep Bob Jenson). “Extirpation” refers to extinction of a species in part of its range. See Michael Allaby, A Dictionary of Ecology 145 (4th ed 2010) (“extirpation” means “[t]he bringing of a species to extinction within a part of its range”).
Petitioner generally agrees with the above plain-text reading of the statute, but it argues for a more demanding “technical” meaning of “persistence” found in the Oregon Plan. The difficultly with petitioner’s argument is that the Oregon Plan does not define “persistence” or any other similar term, and petitioner does not offer a technical meaning that it believes is contained in the Oregon Plan.11 Rather, the Oregon Plan sets out a general mission and goals to, among other things, aid in the recovery of listed species to sustainable population levels. See, e.g., ORS 541.898(2)(a) (“The mission of the Oregon Plan is to restore the watersheds of *210Oregon and to recover the fish and wildlife populations of those watersheds to productive and sustainable levels in a manner that provides substantial ecological, cultural and economic benefits.”); ORS 541.898(5) (“The purpose of the Oregon Plan is to enhance, restore and protect Oregon’s native salmonid populations, watersheds, fish and wildlife habitat and water quality, while sustaining a healthy economy.”); ORS 541.898(7) (“The Oregon Plan shall focus on aiding the recovery of species listed as threatened or endangered under the federal Endangered Species Act or under ORS 496.171 to 496.192 until such time as recovery is achieved.”); ORS 541.890(9) (“Restore” or “restoration” means “to take actions likely to achieve sustainable population levels of native fish or wildlife and their habitats.”); ORS 541.898(l)(e) (With respect to salmonid recovery, “recovery” means “that a proportion of the constituent populations of naturally produced native fish belonging to a listed unit are sufficiently abundant, productive and diverse in life histories and distribution such that the listed unit as a whole is likely to be self-sustaining into the foreseeable future.”). The mission and goals in the Oregon Plan do not contradict the legislature’s understanding of persistence as stated on the Senate and House floors to mean “a forecast of future population health stated in terms of the probability of extirpation.” Tape Recording, Senate Floor Debate, HB 3038, June 15, 2005, Tape 186, Side B (statement of Sen Charlie Ringo).
The legislative history of ORS 537.230(2)(c) thus emphasizes that, in using the words “maintain * * * the persistence of fish species,” the legislature focused on the long-term preservation or endurance of fish population health in the affected waterway. That understanding is consistent with the plain language of the statute, and, thus, we conclude that the legislative intent expressed in ORS 537.230(2)(c) is that, for extension requests subject to that subsection, the department must find that the undeveloped portions of the permits are subject to conditions that preserve from decline the continued existence, or endurance, of listed fish species in the affected waterway.
In the final orders, the department, based on ODFW’s advice, focused on what streamflow is needed to maintain the listed fish populations in the long term, *211even though some short-term negative effects may occur to habitat and individual fish in low-flow years. Petitioner, however, takes issue with what it views as the department’s overstated focus on long-term fish persistence. That emphasis is improper, petitioner argues, because the statute does not allow for any decline in persistence, even if short term. In conjunction with that argument, petitioner sets out three specific ways in which it asserts that the department’s final orders are contrary to law: (1) The conditions placed on the municipal parties’ permits “would allow the diversions to cause substantial and prolonged incursions into the persistence. flows, impairing habitat and causing some fish to perish.” (2) The department did not base the conditions upon ODFW advice or existing data. (3) The department’s conditions rely on an “eviction” strategy whereby fish persistence is maintained by causing them to leave the affected reach for better habitat, even though the statute requires persistence to be maintained “in the portions of the waterways affected by water use under the permit.” ORS 537.230(2)(c).
We reject each of petitioner’s arguments for two overarching reasons. First, as we explain below, each is necessarily dependent upon a conclusion by us that certain of the department findings are not supported by substantial evidence or that its conclusions are not supported by substantial reason. Petitioner recognizes as much because it relies upon a lengthy “substantial evidence and reason” argument in asserting that the department’s final orders are contrary to law. Those types of challenges, though, require us to apply different review standards than the “legal error” standard that petitioner urges here. Moreover, as explained above, the department did not apply an incorrect interpretation of ORS 537.230 (2)(c). The legislative policy of the statute focuses on long-term fish population health in the affected waterway. It does not express a policy that no habitat may be impaired or that no individual fish may be allowed to perish or leave. The department’s interpretation of the statute contained in its final orders — that the department is required to condition the permits to maintain long-term population health of listed fish species — is consistent with the legislature’s policy.
*212III. SUBSTANTIAL EVIDENCE / SUBSTANTIAL REASON
We begin our review of the department’s final orders with petitioner’s first assignment of error. Petitioner asserts that the department’s fish-persistence finding is not supported by substantial evidence, because it is not based on existing data or the advice of ODFW, is improperly dependent on unknown future actions, and lacks substantial reason. Under ORS 183.482(8)(c), we are required to set aside or remand the department’s final orders if they are not supported by substantial evidence. ORS 536.075(3) (appeal of a final order issued by the department in a contested case is to be conducted according to ORS 183.482, except as otherwise specifically provided).
“Substantial evidence exists to support a finding of fact when the record, viewed as a whole, would permit a reasonable person to make that finding.” ORS 183.482(8)(c). As part of our substantial evidence review, “we [also] look at whether the findings provide ‘substantial reason’ to support the legal conclusion reached by the agency.” Warkentin v. Employment Dept., 245 Or App 128, 134, 261 P3d 72 (2011) (citing Drew v. PSRB, 322 Or 491, 500, 909 P2d 1211 (1996) (“[A]gencies also are required to demonstrate in their opinions the reasoning that leads the agency from the facts that it has found to the conclusions that it draws from those facts.” (Emphases in original.)). Applying that standard of review, we turn to petitioner’s contentions.
A. The department’s additional findings
As a necessary preliminary matter, we first address the 35 additional findings made by the department, which petitioner challenges as unsupported by substantial evidence in the record. In the amended proposed orders, the department made additional findings numbered 32 through 66,12 all of which addressed fish persistence. Those findings *213included the department’s ultimate findings regarding what was necessary for fish persistence in the affected reach, and addressed petitioner’s fish-persistence evidence and expert’s opinions. Petitioner broadly challenges all of those findings in a single sentence as “a mere ‘recitation of evidence, followed by a bare conclusion.’” That challenge is not specific enough for us to conduct a review of the findings for substantial evidence or reason. Beall Transport Equipment Co. v. Southern Pacific, 186 Or App 696, 700 n 2, 64 P3d 1193, adh’d to as clarified on recons, 187 Or App 472, 68 P3d 259 (2003) (“[I]t is not this court’s function to speculate as to what a party’s argument might be. Nor is it our proper function to make or develop a party’s argument when that party has not endeavored to do so itself.”).
Petitioner, however, does make specific challenges to findings 43 [59] [61], 45 [61] [63], 46 [62] [64], 48 [64] [66], 49 [65] [67], 63 [79] [81], 64 [80] [82], 65 [81] [83], and 66 [82] [84] as unsupported by substantial evidence in the record based on some cited record evidence. Having reviewed those findings, in view of the whole record, we conclude that they are supported by substantial evidence. We address other specific additional findings of the department as they pertain to petitioner’s more developed arguments below.
B. Fish-persistence flows
Many of petitioner’s contentions appear to stem from petitioner’s initial premise that the persistence flows identified in the final orders must be met each year and in each season to maintain fish species persistence. Thus, petitioner argues, the failure of the conditions to ensure that flows are met in each season means that the department’s fish-persistence finding is unsupported by substantial evidence. However, in its reply brief, petitioner retreats from that position and “agrees [with the municipal parties that] the flows were ‘intended to function as triggers for determining what actions might be necessary when flows fell below those targets’; the clear problem is that, during summer, they do not trigger anything (including ODFW’s intended actions). *214Outside of summer, they trigger inadequate actions.” Upon review of ODFW’s advice and the department’s orders, we agree with petitioner’s refined position that, based on ODFW’s advice, the department set the persistence flows as target levels used to trigger actions required by the permit conditions, and not as hard numbers that must be met at all times.
Our task, thus, is to determine (1) whether substantial evidence supports the department’s findings regarding the effect on the persistence of the listed fish species when the persistence flows are not being met and (2) whether, in turn, substantial reason supports the department’s conclusions that its permit conditions will maintain the persistence of the listed fish species in the affected reach with those effects in mind. With our broad-view task thus defined, we turn to petitioner’s specific contentions regarding the persistence flows.
Petitioner’s main contention is that substantial evidence does not support the department’s findings with regard to the effect of drops below the persistence flows, particularly during historically low-flow months (July 1 to early October), and that substantial reason does not support the department’s conclusion that the permit conditions will maintain the persistence of listed fish species because those conditions do not protect persistence flows over the long term. Specifically, petitioner asserts that persistence flows are not protected because (1) the permit conditions do not limit any drops in flow to the “short-term,” which is not defined in the orders; (2) its expert predicts that the flows will significantly fall below targets for long durations, particularly during July through October; and (3) even if drops in flow were limited to short-term drops, finding 38 [54] [56] that “[t]he short-term drops *** are not incompatible with maintaining the persistence of listed fish species” is not supported by substantial evidence.
With the exception of finding 38 [54] [56], petitioner does not identify the specific findings that it challenges as lacking substantial evidence with regard to streamflow. However, the department made the following relevant findings:
*215“35. ODFW’s recommended streamflows are required on a long-term basis to maintain the persistence of listed fish species in the lower 3.1 miles of the Clackamas River.
“36. ODFW’s recommended minimum streamflows are not presently met on some occasions during the months of July, August, September and early October.
“37. Listed fish species presently tolerate short-term streamflows below the minimum recommended stream-flows, and will likely continue to do so.
“38. The short-term drops below minimum streamflows predicted by Jonathan Rhodes are not incompatible with maintaining the persistence of listed fish species.
“39. The lower 3.1 miles of the Clackamas River represent less than 2% of the available rearing habitat in the Clackamas River basin, and is the least desirable rearing habitat within the basin.
* * * *
“44. Streamflow in the lower 3.1 miles of the Clackamas River during the period April through June is typically ‘well over’ (typically more than 200 cubic feet per second above) the minimum streamflow values recommended by ODFW.
“45. A fish count conducted at sites in the lower 3.1 miles of the Clackamas River in August and early September of 2008 and 2009 found small numbers of steelhead and Chinook.
“46. The lower 3.1 miles of the Clackamas River is likely to be ‘avoided by most species of concern during the warmest time periods in July and August.’
“47. Reducing streamflows below levels typically experienced in the lower 3.1 miles of the Clackamas River during the later part of the summer may cause certain individual fish to either leave this reach of river to find better habitat, or be unable to do so and not survive.
“48. The use of Timothy Lake releases that are available to the permit holders will not always be sufficient to raise streamflows in the lower 3.1 miles of the Clackamas River to the minimum streamflows recommended by ODFW. ODFW’s advice acknowledges this fact and ODFW took *216this into account when concurring with [the department’s] fish persistence conditions.
“49. The amount of water available to the permit holders from Timothy Lake under an agreement with Portland General Electric will vary from year to year. In some years there may not be any water available to the permit holders under this agreement. ODFW is aware of this fact and took it into account when concurring with [the department’s] fish persistence conditions.
“50. ODFW intended the strategy resulting from the annual meeting between the municipalities and ODFW to be documented in writing.
“51. ODFW intended that the municipalities and ODFW will reach mutual agreement on an annual strategy to maximize fishery benefits from any available releases of stored water from Timothy Lake; however, ODFW intends to devise the strategy itself if the municipalities and ODFW are unable to reach agreement on a strategy after good faith effort.
“52. ODFW intended that the annual meeting may cover issues other than Timothy Lake releases that are relevant to both use under the permits and to listed fish species; however, ODFW intends that the strategy include actions pertaining to such issues only upon mutual agreement by ODFW and the municipalities.
“53. Timothy Lake sits roughly 23 miles upstream from the lower 3.1 miles of the Clackamas River.
“54. Releases of water from Timothy Lake affect the entire reach of the Clackamas River downstream from the Lake, and not just the lower 3.1 miles of the River.
“55. The timing and manner of releases from Timothy Lake can have detrimental effects on listed fish species. A release of Timothy Lake water, followed by a poorly timed shut-off of that release, could dewater spawning areas and strand fish for the entire reach of Clackamas River downstream of Timothy Lake.
“56. ODFW’s fish persistence advice is based upon persistence of listed species in the lower 3.1 miles of the Clackamas River, and does not reflect fish flow needs further up the Basin.
*217“57. During the summer months, most of the habitat available to maintain the listed fish species is upstream from the lower 3.1 miles of the Clackamas River.
“58. During the period from the first Monday in September through June 30th, if the minimum fish persistence flows are not met, the municipalities must reduce their diversions by the percentage by which the fish persistence flows are not being met, based on a seven-day rolling average of mean daily flows (e.g., if the fish persistence flows are being missed by 10%, the municipalities must reduce their diversion under the undeveloped portions of the permits by 10% from the maximum amount legally permitted).
“59. During the period from July 1st through the day prior to the first Monday in September, [the department’s] fish persistence conditions permit continued diversion of the undeveloped portions of the permit when the recommended streamflows are not being met.”
(Record citations omitted.)
Based on its additional findings, the department explained that ODFW recognized that the listed fish species have persisted under current conditions where flows are not always met from July to October and that “ODFW has stated that the target flows are what are required on a long-term, rather than short-term basis for persistence of listed fish species. *** ODFW believes that the fish persistence conditions are sufficient to mitigate for the additional diversions contemplated under the permits.” (Emphasis in original.) The department also stated that there was no evidence in the record that the potential movement of fish from the lower reach or the loss of individual fish in the summer months “poses a threat to the persistence of any listed fish species” (Emphasis in original.) The department thus concluded that the permits, as conditioned, maintained the persistence of listed fish species in the affected waterway.
Our main difficulty with petitioner’s omnibus substantial evidence challenge to the department’s findings is that petitioner does not identify problems with particular findings and does not engage with the administrative record as a whole, relying instead upon its own expert, Rhodes. “In cases in which expert witnesses testify on both sides of a *218contested issue, our function is not to weigh the evidence anew to determine which we find more persuasive; it is to determine only whether a reasonable person could weigh the evidence as the finder of fact did.” Kniss v. PERB, 184 Or App 47, 52, 55 P3d 526 (2002) (citing Armstrong v. AstenHill Co., 90 Or App 200, 206, 752 P2d 312 (1988)). Based on review of the record as a whole, we conclude that the department could reasonably weigh the expert evidence as it did and that the department’s findings set forth above are supported by substantial evidence, with the exception of finding 38 [54] [56].
With respect to finding 38 [54] [56], there is not substantial evidence in the record as to what a “short-term drop” is in terms of fish persistence or why the flows predicted by Rhodes, which are not identified in the order, fall within that category. The only evidence pointed to in the record by respondents is testimony from ODFW employee Kepler, which, with respect to that finding, reduces to his bare statement that “Mr. Rhodes’ testimony does not provide information that would alter [ODFW’s] assessment.” Bare conclusions by agency experts cannot be used as a substitute for evidence presented at a contested case hearing. See Drew, 322 Or at 498-99 (“Tt is one thing *** to say that an agency may employ its experience and expertise to evaluate and understand evidence and quite another to allow it to use its special knowledge as a substitute for evidence presented at a hearing.’ ” (quoting Rolfe v. Psychiatric Security Review Board, 53 Or App 941, 951, 633 P2d 846, rev den, 292 Or 334 (1981) (alteration in Drew))).
We also agree with petitioner that the department’s ultimate determination that the permits, as conditioned, will maintain the persistence of listed fish species lacks substantial reason because the department failed to explain how its streamflow findings and the imposed conditions connect, so as to reach that determination. We turn now to that argument.
To understand the problem with the department’s ultimate determination, it is necessary for us to first describe what the permit conditions both do and fail to do to protect the persistence of fish species. As set forth above, 268 Or *219App at 202-03, the department placed three conditions on each of the municipal parties’ permits. The first condition merely adopted ODFW’s recommended persistence flows and designated the location and gage at which those flows would be measured. The second condition is the “meeting condition,” which provides:
“b. In cooperation with the holders of Permits S-46120, S-35297, S-43170, S-22581, S-3778, S-9982, S-32410 and S-37839, the permittee must have an annual meeting with ODFW to devise a strategy to maximize fishery benefits that can be derived from the agreement with PGE for the release of stored water from Timothy Lake. It is [the department’s] intent that ODFW and the permittees shall reach agreement on the strategy. However, if after making a good faith effort ODFW and the permittees are unable to reach agreement on a strategy, ODFW shall devise the strategy. In either case, the strategy shall be documented in writing and the permittees shall comply with the strategy. The annual meeting and resulting strategy may cover issues other than Timothy Lake releases that are relevant to both use under Permits S-46120, S-35297, S-43170, S-22581, S-3778, S-9982, S-32410 and S-37839 and to listed fish species; however, the strategy may include actions pertaining to such issues only upon mutual agreement by ODFW and the permittees.”
Petitioner points out, and respondents do not dispute, that the meeting condition “does not actually require any water to be released from Timothy Lake and specifies that any non-Timothy Lake actions are strictly voluntary.” The department found that Timothy Lake releases may not be available every year or in an amount that will meet persistence flows. ODFW’s advice focused on using releases in the fall months to provide steady flow to protect Chinook spawning and migration, and ODFW would only want releases during the summer if available and if it would not threaten steady flows in the fall.
The third condition is the “curtailment condition,” which provides, in relevant part: “[f]rom the first Monday in September through June 30 the maximum total amount of the undeveloped portion of the [permit] that can legally be diverted shall be reduced in proportion to the amount by *220which the flows shown in Table 1 are not met[.]” As found by the department, the municipal parties are not subject to curtailment of diversions of the undeveloped portions of their permits during the summer, when persistence flows are anticipated to be missed. And, as pointed out by petitioner, and undisputed by respondents, the curtailment condition does not ensure that persistence flow levels will be met or that actual diversions will be reduced, because the proportional curtailment may only require the municipal party to not divert an amount of water already not being diverted under the undeveloped portion of the permit.
Based on the record, petitioner is correct that the conditions do not ensure that persistence flows will ever be met during the summer months, because curtailment by the municipal parties is not required and augmentation flows are typically not to be used in the summer (assuming augmentation flow is available). Petitioner is also correct that the conditions may not ensure that persistence flows are met during the critical fall months. That is because the curtailment condition does not require persistence flows to be met or require actual diversions to be reduced (the municipal parties can “curtail” use by applying the required curtailment to a portion of the permit not being diverted), and augmentation flows may not be available. What we must determine, then, is whether, in light of those circumstances, the department’s finding that the permits, as conditioned, maintain the persistence of listed fish species is supported by substantial evidence and substantial reason.
Petitioner essentially argues that those circumstances necessarily mean that the department’s determination lacks both substantial evidence and substantial reason because the department also found (finding #35) that “ODFW’s recommended streamflows are required on a long-term basis to maintain the persistence of listed fish species in the lower 3.1 miles of the Clackamas River.” In sum, petitioner argues that “[the department] fails to explain how permit conditions with no mechanism to protect persistence flows from July 1 until early September (finding #59), and which only require a modest proportional reduction for the remainder of the year (finding #58), will maintain persistence flows in the long-term as the [final order] finds *221is necessary to maintain fish persistence (finding #35).” (Emphasis in original.)
Respondents provide a three-part response to that argument. They assert that the department and ODFW found that any dips below the persistence flows would not threaten the persistence of the listed fish species. They also contend that petitioner’s expert’s opinion — that missed flows would occur more often — was based on a flawed analysis and was rejected by the department. Respondents also argue that the department did fully explain how the conditions will maintain persistence flows, quoting the following passage from the amended proposed orders:
“In addition, ODFW’s advice contemplates that the target flows will not always be met during the July through October period. Indeed, the advice recognizes that they are not always met presently, and that the listed species have persisted under these conditions. Rather, ODFW has stated that the target flows are what are required on a long-term, rather than short-term basis for persistence of listed fish species. ODFW believes that the fish persistence conditions are sufficient to mitigate for the additional diversions contemplated under the permits. Mr. Kepler testified that the short-term drops below target flows predicted by Mr. Rhodes are not incompatible with maintaining the persistence of listed fish species. WaterWatch provided no evidence to the contrary.”
(Record citations omitted; emphasis in original.)
We agree that the above paragraph is the best explanation provided by the department to connect its findings to its conclusion that it has satisfied ORS 537.230(2)(c) in relation to the substantial reason problem identified by petitioner. However, we conclude that it is not sufficient.
As set out above, the department made certain crucial findings about the persistence flows:
“35. ODFW’s recommended streamflows are required on a long-term basis to maintain the persistence of listed fish species in the lower 3.1 miles of the Clackamas River.
“36. ODFW’s recommended minimum streamflows are not presently met on some occasions during the months of July, August, September and early October.
*222“37. Listed fish species presently tolerate short-term streamflows below the minimum recommended stream-flows, and will likely continue to do so.
“38. The short-term drops below minimum streamflows predicted by Jonathan Rhodes are not incompatible with maintaining the persistence of listed fish species.”
Thus, the department emphasized that the short-term failure to meet persistence flows would not prevent the persistence of the listed fish species, but that meeting those persistence flows on a long-term basis is necessary for fish persistence. The department’s findings also necessitate that the department implicitly found that the permits, without conditions, would contribute to the long-term missing of the persistence flows.13 ODFW’s advice and testimony supports those findings. The department and ODFW, however, did not present evidence of or adequately explain what a “short-term” drop in flow means versus what maintaining “long-term” flows means in terms of fish persistence — that is, the department in its final order glosses over the dispute about when missing the persistence-flow mínimums adversely affects the persistence of the listed fish populations. Is it strictly a durational flow issue or is it related to severity as well? Is missing persistence flows from July through early October in perpetuity a short-term or long-term drop? Without those missing connectors in the final orders, we cannot discern why the department, in reliance on ODFW’s bare conclusion, concluded that the conditions are sufficient, because, contrary to respondents’ arguments, the department did not find, and ODFW did not testify, that any drops in flows do not threaten persistence nor that any long-term drops would never occur.
The department’s orders also do not explain how the municipal parties’ diversions of the undeveloped portions of their permits as allowed under the conditions will not contribute to a long-term failure to meet the persistence flows. A reasoned explanation is necessary because the plain wording of the permit conditions allows the municipal *223parties to divert (at least part, if not all of) the undeveloped portion of their permits regardless of whether persistence flows are met and regardless of how long the failure to meet those flows persists. The department failed to connect the dots between its finding of what is necessary to maintain fish persistence — long-term meeting of persistence flows— with how the conditions ensure that the diversion of the undeveloped portions of the municipal parties’ permits do not contribute to the long-term failure to meet persistence flows.
That missing connection between the what and the how is particularly needed here because the meeting condition, on which the department (based on ODFW’s advice) appears to have particularly relied as the means by which fish persistence will be maintained, is not a condition placed on the use of the undeveloped portion of the municipal parties’ permits. Although the department may have been hopeful, or even confident, that the municipal parties and ODFW will agree on a strategy each and every year to ensure that diversion of the municipal parties’ undeveloped portions of their permits do not prevent persistence flows from being met on a long-term basis, the statute requires more. The statute requires the department to find that “the undeveloped portion of the permit is conditioned to maintain * * * the persistence of [listed] fish species.” ORS 537.230(2)(c) (emphasis added). As written, the department’s meeting condition is not tied to the municipal parties’ diversion of the undeveloped portions of their permits and, indeed, prohibits the municipal parties from being subjected to a reduction in their diversions as part of the annual strategy, absent their voluntary agreement to do so. The department’s findings regarding ODFW’s intent with regard to the meeting condition cannot relieve the department of its statutory obligation to find that the undeveloped portions of the municipal parties’ permits are conditioned, i.e., their use is made conditional, on maintaining fish persistence.
The explanation provided by the department to support its conclusion that the permit conditions meet the statutory standard under ORS 537.230(2)(c) is that, because ODFW “concurs” (with no other explanation) with the department’s finding that the conditions maintain *224fish persistence, then, therefore, the conditions maintain fish persistence.14 The department’s reliance on ODFW’s one-sentence “concurrence” as providing substantial evidence that the department fulfilled its statutory duty is misplaced. That concurrence was provided before the contested case hearing in this matter, and, thus, before the submission and consideration by the department of additional “existing data” and before the department significantly changed the meeting condition, after both the ALJ and the department found that that condition was not, in fact, consistent with ODFW’s advice. Given the department’s own finding, the timing of ODFW’s concurrence, and the nature of that concurrence (a one-sentence conclusory statement), the concurrence cannot constitute substantial evidence that the department’s finding, based on modified conditions and significant additional findings, “will result in the maintenance of the persistence of listed fish species.” The department’s use of ODFW’s “concurrence” and circular reasoning as a substitute for evidence and an explanation that connects its findings to its conclusions, does not pass muster under our substantial evidence review. See Drew, 322 Or at 498-99 (“It is one thing * * * to say that an agency may employ its experience and expertise to evaluate and understand evidence and quite another to allow it to use its special knowledge as a substitute for evidence presented at a hearing.”); id. at 500 (stating that “agencies also are required to demonstrate in their opinions the reasoning that leads the agency from the *225facts that it has found to the conclusions that it draws from those facts” (emphases in original)).
In sum, we conclude that the department’s finding 38 [54] [56] lacks substantial evidence and that the department’s determination that ORS 537.230(2)(c) has been satisfied lacks substantial evidence and substantial reason. Because our conclusion requires the department to revisit its fish-persistence findings and conditions, we do not address petitioner’s contentions that the conditions are not consistent with ODFW’s advice and improperly rely on future conditions.15
IV. MODIFICATION OF ALJ’S FINDINGS OF HISTORICAL FACT
Petitioner next challenges the department’s modification of the ALJ’s recommended annual-meeting condition, asserting that it was an erroneous modification of a finding of historical fact. That challenge is governed by ORS 183.650(3) and (4) (2007), amended by, Or Laws 2009, ch 866, § 7,16 which provide:
“(3) An agency conducting a contested case hearing may modify a finding of historical fact made by the administrative law judge assigned from the Office of Administrative Hearings only if the agency determines that the finding of historical fact made by the administrative law judge is not supported by a preponderance of the evidence in the record. For the purposes of this section, an administrative law judge makes a finding of historical fact if the administrative law judge determines that an event did or did not occur in the past or that a circumstance or status did or did not exist either before the hearing or at the time of the hearing.
“(4) If a party seeks judicial review of an agency’s modification of a finding of historical fact under subsection (3) *226of this section, the court shall make an independent finding of the fact in dispute by conducting a review de novo of the record viewed as a whole. If the court decides that the agency erred in modifying the finding of historical fact made by the administrative law judge, the court shall remand the matter to the agency for entry of an order consistent with the court’s judgment.”
We have explained that, in reviewing de novo under subsection (4), “we apply a ‘preponderance of evidence’ standard.” Corcoran v. Board of Nursing, 197 Or App 517, 525, 107 P3d 627 (2005). In invoking our de novo review function, the petitioner “must specifically identify each challenged modification of a finding of historical fact and explain why that modification was erroneous as unsupported by a preponderance of the evidence.” Id. at 526 (emphases in original). If the petitioner fails to undertake that exercise, we will not engage in de novo review of the challenged modified fact. See Weldon v. Bd. of Lic. Pro. Counselors and Therapists, 266 Or App 52, 64, 337 P3d 911 (2014) (so explaining under current version of ORS 183.650).
We begin with a detailed account of petitioner’s assertions, because respondents argue that petitioner has not sufficiently heeded our instruction in Corcoran to invoke our de novo review. Petitioner asserts that “the ALJ made several findings of fact regarding what ODFW’s advice required from the annual meetings and where the permit conditions failed to incorporate ODFW’s advice.” Specifically, petitioner identifies the following findings of historical fact made by the ALJ:
“The ALJ found that ODFW’s advice intended the annual meetings to address more than just possible releases from Timothy Lake. Specifically, the ALJ found that ODFW’s advice intended that the ‘annual meetings [would] address how to respond to any shortfalls in the target flows,’ and that ‘[the meetings] are, in essence, an opportunity for ODFW, [the department] and the municipalities to work out the competing water needs and interest in light of current water conditions and availability.’
“Critically, the ALJ found, based on testimony at the hearing, that in the event ODFW and the municipalities could not reach agreement on any of the issues addressed *227during the annual meetings, ODFW intended that the permit conditions would ‘require the municipalities to accede to ODFW’s fish persistence standards.’”
(Record citations omitted; brackets and emphasis in original.)
Petitioner then asserts that the department modified those findings with its additional finding of fact number 52 [68] [70], which provides:
“ODFW intended that the annual meeting may cover issues other than Timothy Lake releases that are relevant to both use under the permits and to listed fish species; however, ODFW intends that the strategy include actions pertaining to such issues only upon mutual agreement by ODFW and the municipalities”
(Emphasis added by petitioner.) Petitioner argues that that modification is not supported by the record because the testimony of an ODFW employee, Kepler, demonstrates that “ODFW intended that the outcome of the annual meetings would require more than voluntary water reduction by the permit holders if persistence flows were not being met during the summer.”
We cannot agree with respondents that petitioner has not heeded our instruction in Corcoran. Petitioner has sufficiently identified the historical fact that it alleges the department modified — that ODFW intended that the municipal parties would be required to accede to ODFW’s fish-persistence standards if agreement cannot be reached at the annual meeting — and explained, with record citations, why it believes the modification is not supported by a preponderance of evidence in the record.
However, we are unable to engage in the de novo review that petitioner requests for a different reason. Petitioner has not identified a “historical finding of fact” made by the ALJ that the department has modified, although it has done its best to craft one through selective quotation of the ALJ’s proposed order. The portion of the ALJ’s proposed order on which petitioner relies does not in fact contain a finding of historical fact; rather, it contains a best practices recommendation to the department that the *228ALJ believed was important. The full quotation of the relevant portion of the ALJ’s opinion is as follows:
“Second, although the conditions are written to require a meeting in which ODFW and the municipality agree as to what should be done in a given year, the condition should be clarified to address the situations in which ODFW and the municipality are not able to reach an agreement.
“It is clear that situations could arise in the coming years where the interests of ODFW and the municipalities could diverge. Although the concept of consensus is an important one and it is to be hoped that the annual meeting will lead to a unified plan for each and every year, WaterWatch correctly argues that the PFOs should have a provision for what to do when the parties disagree. Based upon the importance the Legislature has placed on maintaining the persistence of listed fish, the PFO conditions should be clarified to require the municipalities to accede to ODFW’s fish persistence standards if agreement cannot be reached.”
Consistent with that discussion, the ALJ concluded its proposed order with the recommendation that the department modify its PFOs to “include a provision addressing how to resolve situations where ODFW and the municipality cannot agree on all factors at the annual meeting.” Nothing in the sections of the proposed order to which petitioner points contains a finding of historical fact of ODFW’s intent with regard to municipal parties acceding to ODFW in the event of disagreement on issues not related to Timothy Lake releases. We also note, in response to certain arguments petitioner made on reply and in oral argument on this point, that ODFW’s advice to the department also did not contain any such requirement. See 268 Or App at 196-98 (quoting ODFW’s advice).
Because petitioner has not identified a finding of historical fact of the ALJ to which ORS 183.650(3) (2007) applies, we are not permitted to undertake de novo review as provided in ORS 183.650(4) (2007).
V. PROCEDURAL CHALLENGES
Finally, petitioner makes two separate procedural challenges to the department’s final orders. We review procedural challenges to determine if “the fairness of the *229proceedings or the correctness of the action may have been impaired by a material error in procedure or a failure to follow prescribed procedure, including a failure by the presiding officer to comply with the requirements of ORS 183.417(8).” ORS 183.482(7). ORS 183.417(8) requires the ALJ to “ensure that the record developed at the hearing shows a full and fair inquiry into the facts necessary for consideration of all issues properly before the presiding officer in the case and the correct application of the law to those facts.”
Petitioner’s first procedural challenge is difficult to deconstruct, but we understand petitioner to argue that the ALJ violated ORS 183.417(8) by denying petitioner the opportunity to have a neutral fact finder (the ALJ) make findings of fact with regard to its fish-persistence evidence. Petitioner reaches that conclusion through two steps: First, the ALJ accepted an “instruction” from the department that its role in determining the department’s compliance with the fish-persistence requirement in ORS 537.230(2)(c) was limited to determining if the department’s findings were consistent with ODFW’s advice and, thus, the ALJ did not make findings on petitioner’s fish-persistence evidence. Second, the department then made its own independent findings on petitioner’s fish-persistence evidence after concluding in the amended proposed orders that the ALJ’s construction of ORS 537.230(2)(c) was not correct, thus circumventing possible de novo review by this court under ORS 183.650(4) for modifications of historical fact.
We reject petitioner’s assertion. Petitioner’s argument has nothing to do with whether the ALJ failed to ensure a full and fair development of the factual record necessary for a decision in these cases, which is what ORS 183.417(8) requires. Rather, petitioner’s argument is solely directed at its disagreement with the ALJ’s legal conclusion of what ORS 537.230 (2)(c) required of the department, which affected the substance of the ALJ’s findings. The department was permitted to make substantial modifications to the department’s proposed order — such as correcting the ALJ’s incorrect legal conclusion and making additional necessary findings of fact — as long as those modifications were identified and explained. ORS 183.650(2). There is nothing in the *230record that indicates that the ALJ was improperly accepting “instruction” from the department to use an improper construction of the statute with the aim of preventing consideration of petitioner’s evidence. The ALJ explained in the proposed orders that his legal conclusion was based on his reading of the plain wording of the statute and what it required the department to do — that the department “shall” base its findings on ODFW’s advice. Also, except as asserted in petitioner’s second procedural challenge, petitioner does not point to any procedural error committed by the ALJ or the department.
Petitioner’s second challenge contains both a procedural and evidentiary component and pertains to the ALJ’s and the department’s handling of its climate-change exhibits and the DEQ letter exhibits. As discussed in more detail above, 268 Or App at 199-200, the ALJ excluded those exhibits at the hearing as irrelevant, and petitioner made offers of proof of those exhibits. In its amended proposed orders, the department concluded that the climate-change exhibits were relevant and made findings of fact with respect to them, but adhered to the ALJ’s exclusion of the DEQ letters. On judicial review, petitioner argues that (1) the department abused its discretion in admitting petitioner’s climate-change exhibits without first remanding to the ALJ for further hearings as required by OAR 137-003-0655(5),17 and (2) petitioner’s DEQ letter exhibits were unlawfully excluded because they were relevant to fish persistence.
With regard to the asserted procedural error, even if the department violated OAR 137-003-0655(5), which we do not decide, petitioner makes no effort to explain why that error necessitates a remand to the department, such as identifying how the record would have been developed differently absent the error. We may remand an agency final order for *231procedural error only if the agency’s failure to follow procedure impaired “the fairness of the proceedings or the correctness of the action.” ORS 183.482(7). Petitioner has not made that showing. See Gleason v. Oregon Racing Comm., 233 Or App 164, 169, 225 P3d 123 (2010) (“[I]n the absence of a showing that the delay compromised petitioner’s ability to have a fair hearing, we conclude that the procedural deficiency does not require a reversal of the commission’s order.”).
Petitioner’s challenge to the exclusion of the DEQ letters fails for the same reason. We may reverse based on evidentiary error only if the error substantially prejudiced petitioner’s rights. ORS 183.450(1) (“[E]rroneous rulings on evidence shall not preclude agency action on the record unless shown to have substantially prejudiced the rights of a party.”); Adams v. Board of Clinical Social Workers, 201 Or App 500, 507, 119 P3d 260, rev den, 339 Or 700 (2005) (“[W]e conclude that any error in excluding the evidence did not substantially prejudice petitioner and is not, for that reason, a basis for reversal.”). Petitioner’s only attempt to identify prejudice is to assert on reply that it was not able to cross-examine witnesses about the DEQ letters or rely on them in briefing. Petitioner, however, makes no attempt to explain why that inability substantially prejudiced its rights, particularly in light of petitioner’s concession that its expert was permitted to testify about at least one of the three DEQ letters. Accordingly, we reject petitioner’s procedural and evidentiary challenges.
In sum, we conclude that finding number 38 [54] [56] lacks substantial evidence and that the department’s ultimate determination that the permits, as conditioned, maintain the persistence of listed fish species lacks both substantial evidence and substantial reason. We reject all of petitioner’s remaining challenges to the final orders. Accordingly, we reverse and remand each of the final orders in these three cases for further consideration by the department.
In A148870, reversed and remanded.
In A148872, reversed and remanded.
In A148874, reversed and remanded.
Respondent City of Tigard is not a holder of a water permit at issue in this case. However, Tigard was granted party status below because it has an interest in Lake Oswego’s permits through an intergovernmental agreement with Lake Oswego.
Respondent Sunrise Water Authority jointly holds one permit (S-46120) with the North Clackamas County Water Commission. For ease of reference, we name in the opinion only North Clackamas County Water Commission with respect to that permit.
South Fork’s permits S-3778 and S-9982 have points of diversion that are located in tributaries to the Clackamas River. The department conditioned the grant of extensions of those permits on South Fork obtaining approval to change those points of diversion to a point within the lower 3.1 miles of the Clackamas River, which none of the parties challenge. Thus, for purposes of this opinion, we discuss South Fork’s permits as if those points of diversion already have been relocated.
ORS 537.230(2) provides:
“The holder of a permit for municipal use shall commence and complete the construction of any proposed works within 20 years from the date on which a permit for municipal use is issued under ORS 537.211. The construction must proceed with reasonable diligence and be completed within the time specified in the permit, not to exceed 20 years. However, the department may order and allow an extension of time to complete construction or to perfect a water right beyond the time specified in the permit under the following conditions:
“(a) The holder shows good cause. In determining the extension, the department shall give due weight to the considerations described under ORS 539.010(5) and to whether other governmental requirements relating to the *192project have significantly delayed completion of construction or perfection of the right;
“(b) The extension of time is conditioned to provide that the holder may divert water beyond the maximum rate diverted for beneficial use before the extension only upon approval by the department of a water management and conservation plan; and
“(c) For the first extension issued after June 29, 2005, for a permit for municipal use issued before November 2,1998, the department finds that the undeveloped portion of the permit is conditioned to maintain, in the portions of waterways affected by water use under the permit, the persistence of fish species listed as sensitive, threatened or endangered under state or federal law. The department shall base its finding on existing data and upon the advice of the State Department of Fish and Wildlife. An existing fish protection agreement between the permit holder and a state or federal agency that includes conditions to maintain the persistence of any listed fish species in the affected portion of the waterway is conclusive for purposes of the finding.”
“[North Clackamas] is an intergovernmental agency organized to process safe drinking water from the Clackamas River for its members: [Sunrise Water Authority], the City of Gladstone, and Oak Lodge Water District (OLWD).”
“The cities [of Oregon City and West Linn] formed [South Fork], formerly the South Board Water Commission to administer Permit S-9982 (and their other water rights).”
ODFW’s letters for South Fork’s permits S-3778 and S-9982 differed only in that they also recommended that the extensions be conditioned on South Fork obtaining approval for changes in the points of diversion to a point within the lower 3.1 miles of the Clackamas River.
The department deleted the following phrase with regard to each of the eight permits: “Although the Department understood that it was required to follow the ODFW Advice.” The department explained that it found that the preponderance of the evidence did not support that conclusion, because the evidence cited by the ALJ supported only that ODFW understood that the department was required to follow its advice, not that the department shared that understanding.
Of some relevance here, in the final orders, the department added two findings of fact made by the ALJ that it had inadvertently deleted when it replaced the fish-persistence section of the ALJ’s opinion. Those findings are:
“a. ‘Both agencies [the department and ODFW] agree that the conditions placed in the PFOs were not the verbatim advice given by ODFW.’
“b. ‘There was apparently a miscommunication between the agencies ([the department] and ODFW) and municipalities concerning what would come from those meetings.’”
None of the parties to this judicial review has challenged the department’s findings of what portion of each permit is the “undeveloped portion” subject to the department’s fish-persistence conditions.
Petitioner references a 2006 ODFW memorandum and a 2005 ODFW comment letter on the department’s proposed rules in an effort to give “persistence” a more demanding meaning. However, we reject petitioner’s argument because it seeks to hold the department to an interpretation hy ODFW concerning its role in giving advice to the department, which does not reflect the correct standard of review we must apply. We must determine whether the department’s interpretation and application of ORS 537.230 is consistent with the legislature’s expressed policy, not whether it is consistent with ODFW’s interpretation.
Although the additional findings for each of the contested cases are identical (and in the identical order) the starting number is not. The numbers quoted are from the final order in the Lake Oswego case. The numbers that correspond to the South Fork order start with 48 and continue to 82; the numbers that correspond to North Clackamas order start with 50 and continue to 84. Where we refer to particular numbered findings later in this opinion, we identify them by *213the Lake Oswego number followed in brackets by the South Fork number and then the North Clackamas number, for example, “finding number 32 [48] [50].”
That is so, because the department found, with respect to streamflow, the limiting factor for the fish was the long-term failure to meet persistence flows, and it found that use of the undeveloped portion of the permits, without conditions, would not maintain fish persistence.
The amended proposed order makes clear that that is the basis of the department’s reasoning. In the fish-persistence section, the department explained:
“In the case of these extension applications, [the department] made the required finding that the undeveloped portions of the permits at issue are conditioned to maintain the persistence of listed fish species. [The department’s] finding, and the conditions supporting the finding, is based upon ODFW’s advice pertaining to fish flows needed to maintain the persistence of listed species. ODFW’s advice letter is in turn based upon existing data. [The department’s] finding is therefore based on the two sources of evidence that is permitted — and required' — to consider. ODFW concurred that [the department’s] conditions are consistent with ODFW’s advice. This constitutes substantial evidence that [the department’s] finding and conditions will result in the maintenance of the persistence of listed fish species. As a result, [the department] established a prima facie case showing compliance with ORS 537.230(2)(c).”
(Record citations and footnotes omitted; emphasis added.)
We also reject without discussion petitioner’s two additional arguments related to streamflow — salmon nest desiccation and the gage point location— because petitioner failed to adequately develop those arguments for our review.
Because an ALJ was assigned to these contested cases before August 4, 2009, the effective date of the current version of ORS 183.650(3), the former version applies. Or Laws 2009 ch 866, § 8. The current version requires the agency to find that “there is clear and convincing evidence in the record that the [ALJ’s] finding was wrong.” ORS 183.650(3).
OAR 137-003-0655(5) provides:
“The agency or, if authorized to issue a final order, administrative law judge shall consider any timely exceptions and argument before issuing a final order. If exceptions are received, the agency or the administrative law judge may not consider new or additional evidence unless the agency requests the administrative law judge to conduct further hearings under section (1) of this rule. The agency or administrative law judge may issue an amended proposed order in light of any exceptions or argument.”