1000 FRIENDS OF OREGON, Petitioner,
v.
LAND CONSERVATION AND DEVELOPMENT COMMISSION, Respondent,
Lane County and Oregon Forest Industries Council, Intervenors Respondents.
84-ACK-201; CA A33755.
Court of Appeals of Oregon, In Banc.[*]
Argued and Submitted June 25, 1986. Resubmitted December 3, 1986. Decided January 14, 1987. Reconsideration Denied March 13, 1987.*459 Robert L. Liberty, Portland, argued the cause and filed the briefs, for petitioner.
Michael A. Holstun, Asst. Atty. Gen., Salem, argued the cause, for respondent Land Conservation and Development Comn. With him on the brief were Dave Frohnmayer, Atty. Gen., and James E. Mountain, Jr., Sol. Gen., Salem.
William A. Van Vactor, Eugene, argued the cause and filed the brief, for intervenor-respondent Lane County.
Frank M. Parisi, Portland, waived appearance for intervenor-respondent Oregon Forest Industries Council.
Resubmitted in banc December 3, 1986.
WARDEN, Presiding Judge.
1000 Friends of Oregon (1000 Friends) seeks review of the Land Conservation and Development Commission's (LCDC) acknowledgment of Lane County's Rural Comprehensive Plan. Most of the assignments of error concern portions of the plan which 1000 Friends believes improperly permit the establishment of dwellings on rural agricultural and forest lands. We reverse on all but one of the assignments of error.
In its first assignment, petitioner asserts that provisions of the F-2 (impacted forest lands) zone violate Goal 4. That goal requires that forest lands be conserved for forest uses, which it defines as:
"(1) the production of trees and the processing of forest products; (2) open space, buffers from noise, and visual separation of conflicting uses; (3) watershed protection and wildlife and fisheries habitats; (4) soil protection from wind and water; (5) maintenance of clean air and water; (6) outdoor recreational activities and related support services and wilderness values compatible with these uses; and (7) grazing land for livestock."
None of those uses directly involves residences. However, since Lamb v. Lane County, 7 Or LUBA 137 (1983), LCDC has interpreted Goal 4 to permit dwellings or other non-enumerated uses which are "necessary and accessory" to one of the enumerated forest uses. 7 Or LUBA at 143. Residences permitted under that interpretation are known as forest dwellings.[1] 1000 Friends attacks the conditions on which the county permits forest dwellings, asserting that they are inadequate to preserve forest lands for forest uses. It argues that the conditions violate the goal both as a matter of law and as a matter of fact. We hold that the criteria are legally insufficient to meet the requirements of the goal, as LCDC has interpreted those requirements.[2]
LCDC's requirement that a forest dwelling be necessary and accessory to a forest use[3] is a reasonable construction of the Goal 4 requirement that forest land be preserved for forest uses. LCDC applied that requirement in finding that an earlier version of the county's plan failed to comply with the goal. The question is *460 whether LCDC erred in deciding that the county's current criteria for forest dwellings in the F-2 zone comply. Lane County Development Code § 16.211(3)(b) provides:
"A dwelling or mobile home, and any accessory structures, on a vacant legal lot containing at least 10 acres shall be deemed accessory and necessary to the forest management of the legal lot provided:
"(i) A detailed forest management plan sufficient to obtain a tax deferral is submitted for the legal lot which demonstrates forest production will be enhanced by on-site forest management from the residence, and
"(ii) Based upon the above-referred forest management plan, the property has qualified for a tax deferral pursuant to State law, and
"(iii) The forest management plan shall specify how the following practices, when applicable, are to be addressed: road and fire trail construction and maintenance, site preparation, reforestation, stand conversion, planning of nonstocked openings, competition reduction/release, precommercial thinning, harvest scheduling/rotation and special site treatment for topography and other concerns.
"(iv) If the legal lot does not have a forest deferral pursuant to State law, then upon substantial completion of the details represented in the forest management in LC 16.211(3)(b)(i) above, the dwelling or mobile home shall be allowed on the property. Substantial completion of the details represented in the forest management plan shall be verified by a qualified private forester and such verification shall be submitted in writing to the Department. During the interim, while the forest management plan is being implemented on the subject legal lot, a temporary mobile home in conjunction with the forest use shall be allowed for a period not to exceed 5 years. If the forest management plan is not implemented within the five-year period, the temporary mobile home shall be removed."
Any proposed dwelling which meets the criteria of this provision is automatically "deemed" necessary and accessory to the forest use. Thus the county and LCDC held, as a matter of law, that the criteria are sufficient to assure compliance with Goal 4. In that holding they erred.[4]
The county defines "accessory" as "[i]ncidental, appropriate and subordinate to the main use of a tract or structure." Lane County Development Code § 16.090. That definition is consistent with LCDC's use of the term; LCDC could properly find that any forest dwelling which meets the county's criteria would be accessory to a forest use. The dwelling must also, however, be necessary. Neither LCDC nor the county has defined that term. The dictionary definition is "that cannot be done without: that must be done or had: absolutely required." Webster's Third New International Dictionary 1511 (1976). That *461 definition is compatible with LCDC's use of "necessary"[5] and with Goal 4's requirement that forest lands be preserved for forest uses. Lane County's criteria would allow dwellings which can be done without, need not be had and are not absolutely required for a forest use; they therefore do not comply with the goal.
Lane County Development Code § 16.211(3)(b)(i) requires that there be a forest management plan "which demonstrates [that] forest production will be enhanced by on-site forest management" from the dwelling. (Emphasis supplied.) Many things can enhance production without being necessary to it. It may be more convenient for the owner of forest land to do required cultivation work from a nearby residence rather than commuting from a home some distance away, but that does not make it necessary to do so. Living on the land may help deter arsonists, and thereby enhance production, but that fact does not render a forest dwelling necessary. For a forest dwelling to be necessary and accessory to wood fiber production, it must, at the least, be difficult to manage the land for forest production without the dwelling. The purpose of the dwelling must be to make possible the production of trees which it would not otherwise be physically possible to produce. That requirement, which is inherent in the concept of necessary, is absent from section 16.211(3)(b).[6]
The second assignment attacks Lane County Development Code § 16.211(3)(c), which also applies to the F-2 zone. It permits a
"dwelling or mobile home, and any accessory structures, in conjunction with the propagation or harvesting of a forest product on a vacant legal lot that is managed as part of a woodlot meeting the Douglas-fir cubic foot site index and acreage requirements below, provided sufficient factual documentation concerning the forest management, the soils, cubic foot site indices and acreage of the legal lot is presented by the applicant to verify that the legal lot meets the minimum cubic foot site index acreage requirements specified [in the rest of the section]."
Such "woodlot" dwellings are permitted in exclusive farm use (EFU) zones. ORS 215.213(2)(a), (b). The F-2 zone contains significant small farming units as well as small forested lots, and the county and LCDC argue that it was within LCDC's authority to allow uses permitted in EFU zones in the F-2 zone as well. They do not assert that the F-2 zone is an EFU zone.
Lane County Development Code §§ 16.211(4)(k) and (1) permit woodlot homes on those properties within the F-2 zone that are in predominately agricultural use; 1000 Friends specifically agrees with that provision. It attacks the extension of the "woodlot" dwelling exemption to those portions of the F-2 zone which would not by themselves qualify for EFU zoning. The code does not purport to require that the dwelling be necessary and accessory to a forest use, but only that the land meet certain fertility requirements and be managed for forest production. There is no attempt to show that forest production would require an on-site dwelling in every instance. If LCDC determined that the "woodlot" dwelling provisions meet the requirements of Goal 4 for forest uses, it erred.
LCDC does not appear to claim that it held that the "woodlot" provisions satisfy Goal 4. Rather, relying on Shadybrook Environ. Protec. Assn. v. Wash. Co., 61 *462 Or. App. 474, 658 P.2d 1168, rev. den. 294 Or. 682, 662 P.2d 725 (1982), it argues that it may appropriately determine to allow as conditional uses in mixed forest and farm areas all those uses which are permitted in EFU zones. Its argument is based on a misreading of Shadybrook. The issue in that case was whether a quarry is a permissible non-forest use in a combined agricultural and forestry district. Quarries are permitted without goal exceptions in EFU zones but not in forestry zones. We held that mixed zones have to meet the requirements of both goals. Thus, the statutory exception permitting quarries in EFU zones did not excuse compliance with Goal 4.
However, we also held that it was possible to show that a quarry would not violate Goal 4 and that the county therefore did not have to take a needs exception under Goal 2. We quoted an LCDC policy paper in which the commission stated that criteria for non-forest uses on forest land should, as a minimum, require that the uses be compatible with forest uses, not interfere with forest practices, maintain the overall land use pattern in the area and minimize loss of protective forest lands. We also quoted an opinion in which the commission required that a ski resort be built on land within a forest zone which was not suitable for timber use. 61 Or. App. at 481-482, 658 P.2d 1168. Assuming that "woodlot" dwellings may be non-forest uses, and thus not subject to the necessary and accessory test,[7] nothing in the county's regulations requires that they be built on land not otherwise suitable for timber use or that they meet the other criteria we mentioned in Shadybrook. LCDC does not claim to have relaxed those criteria since that case.[8]
In its third assignment, 1000 Friends attacks the minimum parcel sizes for the F-2 zone. The county based those sizes on the capacity of the parcel to produce $10,000 of wood fiber products per year over the growth cycle. 1000 Friends asserts that the "credible evidence in the record" shows that the parcels are so small that they will encourage residential rather than commercial use of the land. We do not evaluate the credibility of the evidence. See n. 2, supra. We affirm LCDC on this assignment.
The fourth and fifth assignments involve substantial evidence challenges to minimum lot sizes in the F-1 (non-impacted forest land) zone and in the EFU zones. We review acknowledgment orders in the manner that we review orders in contested cases. ORS 197.650(1); ORS 183.482. That manner includes the limitation of our review of the factual findings to a determination of whether there is substantial evidence to support them. ORS 183.482(8)(c).[9] The county cites decisions in which we have held that the substantial evidence test does not allow us to weigh the evidence. See, e.g., Babcock v. Employment Division, 72 Or. App. 486, 488, 696 P.2d 19 (1985). Our cases do not hold, however, that any evidence whatever is *463 substantial. Rather, we look at the nature of the evidence to determine whether it is "such proof as a reasonable mind would employ to support a conclusion." Cook v. Employment Division, 47 Or. App. 437, 441, 614 P.2d 1193 (1980). When the evidence consists of an opinion whose foundation is unclear or which is inconsistent with the information on which it is based, we have held that the evidence was not substantial. Brown v. AFSD, 75 Or. App. 98, 101-102, 705 P.2d 236 (1985), rev. den. 300 Or. 477, 713 P.2d 1058 (1986). That particular evidence was not taken under oath or subject to cross-examination also affects its substantiality. See Kuykendall v. AFSD, 70 Or. App. 526, 530 n. 2, 690 P.2d 517 (1984). We review the evidence in this case with these considerations in mind.
In the fourth assignment, 1000 Friends contends that the county's 40 acre minimum lot size in the F-1 zone violates Goal 4, because it is conducive to residential rather than forestry uses. LCDC relied primarily on the county's findings in support of the 40-acre minimum. Those findings are:
1. Two letters from a representative of the Lane County Land Owners Association stated that 40 acres could be an operationally and economically manageable unit within the timber industry;
2. There are a significant number of parcels within the forest land zone of the 40-acre size;
3. 92 percent of the industrial forestry land which is in parcels of less than 80 acres is in parcels of 40 acres or more, and 25 percent is in parcels of exactly 40 acres (another finding stated the point with somewhat different figures, apparently covering slightly different land; it neither strengthens nor weakens this finding);
4. Commercial forestry representatives stated that 40 acres is a commonly acquired, managed and exchanged unit of commercial forest lands.
In addition to those findings, LCDC also emphasized that the county prohibits new dwellings in the F-1 zone. All of the findings have evidentiary support. However, they do not by themselves constitute substantial evidence for the ultimate finding that a 40-acre minimum is appropriate for forest uses. The letters mentioned in the first finding say what the county states that they say. However, the letters give no specific information to support the conclusion that 40 acres is an economically viable size. They are simply the opinion of one person whose background, basis of knowledge, and personal interests are not in the record. Without this information, neither LCDC nor the county could determine whether it was reasonable to rely on his opinion to support their conclusions. See Cook v. Employment Division, supra. Other information in the study on which the second finding is based strongly suggests that the 40-acre parcels in the F-1 zone are used more for residential than for forest purposes. This finding does not support the conclusion that 40 acres is a proper size for forest uses.
The significance of the third finding is unclear. That most parcels of less than 80 acres are 40 acres or greater says nothing about the overall mix of parcel sizes in the F-1 zone or about the suitability of 40-acre parcels for commercial forest uses. The conclusion appears to have been originally part of the county's attempt to determine the minimum parcel on which it could permit residences in the F-1 zone. It has little relevance to the appropriate minimum parcel size now that the county has decided to prohibit additional residences. The fourth finding is based on the same letters as is the first finding and has the same weaknesses.
The county must justify its parcel size as appropriate for the conservation of forest uses. The second and third findings, in the light of the evidence on which they are based, contribute nothing to the county's conclusion that 40 acres is appropriate. The first and fourth findings, by themselves, do support that conclusion. However, so far as we can determine, they are based entirely on the unsupported opinion *464 of one person.[10] That opinion is insufficient to support the findings.
1000 Friends points out that the county did not analyze the relationship between forest management efficiency and parcel size and that, to the extent that it conducted an inventory of existing parcel sizes, it did not show that the smaller sizes conserve forest land for forest use. If anything, the information which the county did develop suggests a correlation between parcels under 80 acres and increased residential use. There is also no information concerning the size of actual sales between commercial timber owners, particularly when compared with sales for other purposes. In short, there is not substantial evidence in the record to justify the county's decision.[11]
The fifth assignment concerns minimum parcel and farm sizes in the county's EFU zones. The county produced a large amount of data which it used in setting those sizes. No one challenges its accuracy. Rather, 1000 Friends attacks the county's methodology in deciding what information to develop and how to use it to determine the minimums. The issue is thus whether the county's conclusions are reasonably based on relevant facts and resulted in a reasonable method of complying with the Goal 3 requirement that "[a]gricultur[al] lands shall be preserved and maintained for farm use * * *." Determining minimum lot sizes necessarily requires the drawing of inferences and the exercise of judgment. The legislature has given the responsibility for drawing those inferences and exercising that judgment to LCDC. With one major exception, LCDC's actions are reasonable.
Lane County's agricultural minimum lot size provisions, and the method by which the county arrived at them, are complicated. The county established minimum size requirements both for farm operations and for individual farm parcels. The farm operations minimums vary according to the crop, while the individual parcel minimums are different for the different geographic regions within the county. The farm operations minimums range from 20 acres for berries and grapes to 320 acres for the general farm, primarily crops, category. The parcel minimums range from 25 to 60 acres. A farm must meet both the farm operations minimum for its crop and the parcel minimum for its region in order to qualify for a farm dwelling.[12]
1000 Friends does not challenge the farm operations minimums. Its challenge to the parcel minimums focuses on the information the county used in developing them. A crucial foundation for the county's analysis is an inventory of all farm parcels in the county, classified by region. That inventory, together with statements from agricultural extension agents of the smallest practical and the preferred parcel sizes for the different kinds of crops in the county, is the basic information which the county used in establishing minimum parcel sizes. For each region, the county determined the *465 range of parcel sizes for parcels of 10 acres or larger and compared that size to the sizes which the agents recommended for crops which were characteristic of the region. It then adopted a minimum parcel size for that region that was generally both around the 30th or 40th percentile of parcel sizes and close to the agents' recommendations.
If the information on which the county based its inventory reflects actual field sizes, the parcel size determinations were reasonable, and LCDC did not err in approving them. 1000 Friends' objections generally reflect a desire that the county had acted differently rather than a showing that LCDC was incorrect in accepting the county's action.
However, there is one major, glaring flaw in the county's determination. Its inventory of parcel sizes relies entirely on tax lots. Rather than showing that tax lots are "farm units"[13] (parcels worked as one area), it simply assumes that they are. There is no reason in the record to believe that that assumption is correct. A tax lot may be divided into two or more worked areas, or contiguous tax lots may be combined into one area, even if the lots have different owners. Without information about how the lots in each region are actually used, the county cannot know if they reflect existing commercial agriculture. If the county wishes to use this method, it must determine what the relationship between tax lots and actual worked area size is. Without that information, its determination of minimum parcel sizes is built on sand. We reverse on the fifth assignment of error.
The last two assignments require little discussion. The county and LCDC failed adequately to apply Goal 5 to forest land because of LCDC's misunderstanding of the effect of ORS 527.722 and ORS 527.726(1). 1000 Friends of Oregon v. LCDC (Coos County), 79 Or. App. 360, 719 P.2d 65, rev. allowed 301 Or. 445, 723 P.2d 325 (1986); 1000 Friends of Oregon v. LCDC (Tillamook Co.), 76 Or. App. 33, 708 P.2d 370 (1985), on reconsideration 77 Or. App. 599, 714 P.2d 279, rev. allowed 301 Or. 320, 721 P.2d 1388 (1986). We reverse LCDC on the sixth assignment of error.
Finally, 1000 Friends challenges the many committed exceptions the county took. The county described each exception in a two-page form with an accompanying aerial photograph. 1000 Friends does not make a detailed challenge to any particular exception but asserts that the information is insufficient to show that any of the exceptions meets the criteria of ORS 197.732 and the related rules. We will not ourselves undertake the detailed analysis 1000 Friends does not attempt. However, in order to justify a committed exception, the county must discuss how the current land use patterns came about and the relationship between the goals and that development. OAR 660-14-028(2)(c)(A); 1000 Friends of Oregon v. LCDC, supra n. 2, 69 Or. App. at 728, 688 P.2d 103; see also 1000 Friends of Oregon v. LCDC (Curry Co.), 301 Or. 447, 516, 724 P.2d 268 (1986). It has not done that. We reverse LCDC on the seventh assignment of error.[14]
Reversed on the first, second, fourth, fifth, sixth and seventh assignments of error; *466 remanded for further proceedings not inconsistent with this opinion.
NOTES
[*] Joseph, C.J., not participating.
[1] LCDC has also permitted non-forest dwellings if they meet certain criteria. Those criteria are not in issue here, because respondents only defend the county's standards as appropriate for forest dwellings.
[2] It is difficult for us to evaluate 1000 Friends' factual arguments in this case. It does not adequately describe either the county's or LCDC's findings, nor does it, with a few exceptions, attempt to show that there is no substantial evidence in the record to support them. Instead, 1000 Friends sets forth its view of the evidence, generally relying on material it, rather than the county, placed in the record. The most its argument shows is that there is substantial evidence to support the findings that 1000 Friends wishes the county and LCDC had made. That is very different from showing that there is no substantial evidence for the findings which they did make. We have acquiesced in a similar procedure in a case where the county's record and therefore LCDC's was inadequate and the additional facts 1000 Friends presented simply made it more complete. 1000 Friends of Oregon v. LCDC, 69 Or. App. 717, 723-724 n. 5, 688 P.2d 103 (1984). In this case, however, the county's record is, in most respects, much more complete, and 1000 Friends' desired findings would supplant rather than supplement the county's. We do not review LCDC's findings de novo, as 1000 Friends appears to assume.
[3] The county refers to this requirement as 1000 Friends'. It is incorrect. As LCDC recognizes in its brief, the requirement is LCDC's; 1000 Friends simply disputes LCDC's application of it.
[4] Both the county and LCDC urge us to defer to LCDC's construction of Goal 4 in reviewing this order. We agree that some deference is appropriate. See 1000 Friends of Oregon v. Wasco County Court, 299 Or. 344, 369, 703 P.2d 207 (1985). However, giving deference to LCDC does not mean that we should abdicate our responsibility for judicial review. LCDC has interpreted Goal 4 by adopting the "necessary and accessory" test. We accept that test as a reasonable construction of the goal. LCDC has further explained that test for the purposes of this case by accepting the county's definition of "accessory." We also accept that definition. Neither the county nor LCDC attempted to define "necessary" or otherwise explain how that portion of the test applies to this case. We cannot treat LCDC as having implicitly explained its explanation. The mere fact that LCDC acknowledged the county's plan does not further our understanding of its view of the meaning of the goal. If LCDC wishes us to defer to its understanding of the goal, it must articulate that understanding rather than ask us to divine it from its unexplained actions. LCDC on remand may articulate some different meaning for the word or may adopt a different test under Goal 4. We could then review that test or that meaning for its compatability with the goal. If LCDC fails to do so, the responsibility for declaring the meaning of the goal including the meaning of the test LCDC adopted to explain the goal is ours.
[5] In Lamb v. Lane County, supra, 7 OR LUBA at 143, LCDC gave logging roads as an example of what is necessary to a forest use. That example indicates that the forest use must at least be very difficult without the accessory use before the accessory use is necessary. Of course, if LCDC intends "necessary" to have a different meaning, it is free on remand to explain and adopt that meaning.
[6] LCDC argues that it may appropriately approve less restrictive standards for forest dwellings in impacted forest zones than it requires in primary forest zones. However, the only relevant criterion LCDC has articulated for evaluating the standards for forest dwellings in either kind of zone is the necessary and accessory test.
[7] However, LCDC states in its brief: "[N]on-forest dwellings are not an issue in this appeal." That being its position, its reliance on a case involving a non-forest use is puzzling.
[8] In Shadybrook, we also quoted an LCDC policy paper, issued in 1978 and amended in 1979, which stated, with regard to Goals 3 and 4:
"`An exception is not required if the decision involves the use permitted in an EFU Zone under ORS 215.203-.215, is consistent with forest uses defined in Goal 4 or if findings can be made that the land is built on or irrevocably committed to nonfarm or nonforest uses in urban or rural areas.'" 61 Or. App. at 481, 658 P.2d 1168.
The paper does not say that EFU uses are permitted on Goal 4 land. Rather, the first clause describes uses permitted without an exception to Goal 3, the second describes uses permitted without an exception to Goal 4 and the final clause describes what have become known as the built upon and committed exceptions. LCDC and the county have produced no other authority for allowing EFU uses in the non-agricultural portions of mixed forest/agricultural zones.
[9] We review LCDC's order; therefore it is its evidentiary findings which must be supported by substantial evidence. We review the county's findings only to the extent that LCDC relied on them or adopted them. The legislature has not established LCDC's scope of review of the county's factual findings, other than for goal exceptions. ORS 197.732(6)(a).
[10] Evidence in the record, on which 1000 Friends places considerable emphasis, indicates that parcels of less than 66 acres in the F-1 zone will generally be more valuable for residential than for forest uses. In response, LCDC inferred that the county's prohibition of forest dwellings would result in the price of the parcels reflecting forest rather than residential values. That is a reasonable inference which LCDC was entitled to draw. However, the inference simply addresses one of the weaknesses in the county's case for a 40-acre minimum; it does not provide any additional strength to that case.
[11] 1000 Friends also argues that permitting 40-acre parcels will inevitably lead to rezoning to F-2. In contrast to 1000 Friends of Oregon v. LCDC, supra, 69 Or. App. at 733-734, 688 P.2d 103, the record in this case does not affirmatively show that 40-acre parcels are not practical for the appropriate resource use. For that reason, and because our action concerning the F-2 zone may change other relevant factors, we do not consider the argument.
[12] Parcel sizes are particularly important, because a majority of farm operations in Lane County include at least some rented parcels and many include parcels which are not contiguous to each other. Thus, a farm operations minimum by itself could result in widely scattered small parcels that would be inefficient to farm, possibly producing pressures to permit non-agricultural uses.
[13] The county often refers to "farm units" in its discussion. As best we can determine from the record, that phrase encompasses individual farm parcels that is, fields rather than entire farming operations.
[14] We recognize that the large number of exceptions requires, as a practical matter, that the county attempt to justify each one in a summary fashion. However, it must still show compliance with ORS 197.732 and the administrative rules. We also recognize that the large number of exceptions makes it impractical for 1000 Friends to discuss in detail the problems it perceives with individual areas. However, if after remand 1000 Friends continues to believe that some exceptions are improper and again brings the issue to this court, it should at least identify which exceptions it questions and state the general kinds of problems it sees with each. It could do so by lists arranged by categories. That would allow the parties to focus their arguments and would assist us in making our review meaningful. See 1000 Friends of Oregon v. LCDC (Curry Co.), supra, 301 Or. at 518, 724 P.2d 268.