OPINION
BIVINS, Judge.
Ironically, what started out as an application for a one-year variance continues to wind its way through the legal and administrative process and four years later comes before this Court for the second time. Following review of the denial of a variance by the New Mexico Environmental Improvement Board (Board) on the first appeal, we remanded “with instructions to the Board to conduct further proceedings to determine whether the wood smoke, in the volume being emitted from appellant's wigwam burner is ‘injurious to health or safety’.” Duke City Lumber Co. v. N.M. Env. Imp. Bd., 95 N.M. 401, 407, 622 P.2d 709 (Ct.App.1980), cert. denied, 95 N.M. 426, 622 P.2d 1046 (1981) (hereinafter “Duke City I”). Pursuant to that mandate, the Board held a second hearing and again denied the application for a variance filed by Duke City Lumber Co. (Duke City). Duke City appeals.
It raises the following points:
1. Duke City made a prima facie showing that wood smoke, in the volume emitted, is not injurious to health or safety;
2. The Board’s ruling that the Environmental Improvement Division (Division) rebutted Duke City’s showing is not supported by substantial evidence;
3. The Board acted arbitrarily and capriciously in considering undue economic burden and unsworn “citizen testimony”.
I. Standard of Review.
NMSA 1978, § 74-2-9(A) (Repl.Pamp. 1981) provides: “Any person to whom the board denies a variance, after a hearing, may appeal to the court of appeals. All appeals shall be upon the record made at the hearing * * *.” (Emphasis added.) Subsection C of § 74-2-9 provides:
Upon appeal, the court of appeals shall set aside the board’s denial of the variance request only if found to be:
(1) arbitrary, capricious or an abuse of discretion;
(2) not supported by substantial evidence in the record; or
(3) otherwise not in accordance with law.
(Emphasis added.)
The Board in its brief reminds us that on review, “In deciding whether a finding has substantial support, we must view the evidence in the most favorable light to support the finding * * *. Further, only favorable evidence and the inferences to be drawn therefrom, will be considered, and any evidence unfavorable to the findings will not be considered.” United Veterans Org. v. New Mexico Prop. App. Dept., 84 N.M. 114, 500 P.2d 199 (Ct.App.1972). This approach can be contrasted to a review based on the record as a whole. See, e.g., NMSA 1978, § 27-3-4 (Repl.Pamp.1982) (“record as a whole”); New Mexico Human Services Dept. v. Garcia, 94 N.M. 175, 608 P.2d 151 (1980) (review of the whole record); see also NMSA 1978, § 12-8-22(A) (This portion of the Administrative Procedures Act requires consideration of “the entire record”, but only the Human Rights Commission has elected to come under it). The Air Quality Control Act, NMSA 1978, §§ 74-2-1 to 74-2-17 (Repl.Pamp.1981), under which this appeal was taken, provides for review only of whether substantial evidence “in the record” supports the Board’s finding. Section 74-2-9(C). Hence, under the language of that section, we must limit our review to only favorable evidence in support of the Board’s ruling, disregarding any unfavorable evidence. Duke City does not contest this approach but claims this Court may not “completely disregard the deficiencies in the factual evidence upon which the Board relies.”
While we limit our review in this ease to the record, not the record as a whole, we take this opportunity to note some of the shortcomings of this limited review as applied to administrative appeals. First, it would appear that New Mexico’s position is not only outdated, but contrary to the rule followed by other jurisdictions and the federal courts. See Utton, The Use Of The Substantial Evidence Rule to Review Administrative Findings Of Fact In New Mexico, 10 N.M.L.Rev. 103 (1979-80). The. United States Supreme Court in Universal Camera Corp. v. National Labor Rel. Bd., 340 U.S. 474, 71 S.Ct. 456, 95 L.Ed. 456 (1951), discussed the question in depth and held that the Administrative Procedure Act (federal) and the Taft-Hartley Act direct the courts to assume more responsibility for the reasonableness and fairness of N.L. R.B. decisions than some have shown in the past. The Court said, “Congress has merely made it clear that a reviewing court is not barred from setting aside a Board decision when it cannot conscientiously find that the evidence supporting that decision is substantial, when viewed in the light that the record in its entirety furnishes, including the body of evidence opposed to the Board’s view.” 340 U.S. at 488, 71 S.Ct. at 465 (emphasis added). Of course, Universal Camera dealt with a statute requiring review based on the whole record, whereas the Air Quality Control Act provides review as to “substantial evidence in the record.” Section 74-2-9(C). The language, “substantial evidence in the record,” would suggest a legislative intent to prescribe the more narrow review. This assumption is borne out by the broader standard of review required under the Public Assistance Appeals Act, NMSA 1978, §§ 27-3-1 to 27-3-4 (Repl.Pamp.1982), and the New Mexico Administrative Procedures Act, NMSA 1978, §§ 12-8-1 to 12-8-25. It is not clear, however, why the legislature requires different standards of review for different administrative agencies.
A uniform approach to judicial review of administrative determinations could be accomplished by the legislature or perhaps by court interpretation. 10 N.M.L.Rev. at 120. Such an approach has the advantage of facilitating judicial economy as well as promoting even-handed treatment by all state administrative agencies. Aside from uniformity, a review of the entire record is clearly indicated in those cases where the administrative agency serves not only as the fact finder, but also as the complainant and prosecutor. See 73 C.J.S. Public Administrative Bodies and Procedure § 198 (3d ed. 1971). In the present case the role of the Board has been characterized as that of “judge” and the Division as “prosecutor.” While the Division had the responsibility for presenting the State’s case, the Board nonetheless filed a separate brief and vigorously argued the case. In fairness to the Board we point out that a review of the whole record demonstrates that the chairman and Board members did not lose sight of their statutory duty, and we do not direct these observations at this or any other administrative body in particular.
II. The Parameters for Review.
Duke City filed a petition for variance from Air Quality Control Regulation 402(A). Section 74-2-8 permits the Board to grant a variance when it finds (1) that compliance will result in an arbitrary and unreasonable taking of property, or will impose an undue economic burden upon any lawful business; and (2) that the granting of the variance will not result in a condition “injurious to health or safety.”
Duke City did not rely on arbitrary and unreasonable taking of property but instead on undue economic burden. We held in Duke City I that Duke City made a prima facie showing as to this element, and we remanded solely for a determination on the issue of injury to health or safety. Id. 95 N.M. at 404, 622 P.2d 709.
Duke City argues that the three Board members who voted to deny the variance considered economic burden. It says that under the doctrine of law of the case the Board cannot reconsider matters resolved on the first appeal. See Ute Park Summer Homes Ass’n v. Maxwell Land G. Co., 83 N.M. 558, 494 P.2d 971 (1972). We need not address the question of whether or to what extent the law-of-the-case doctrine applies to administrative hearings; the Board resolved the question on the issue of injury to health or safety. That determination alone justifies a denial of the variance, if it withstands the test of appellate review. Section 74-2-9(C). “Undue economic burden” is not an issue in this appeal.
In approaching the health or safety issue we must first determine the degree to which we must answer this question. Both sides acknowledge that it cannot be answered with absolute certainty. Duke City contends, however, that the Division, in rebuttal, must establish injury to health or safety as a medical probability. The Division fakes the position that, in keeping with the statutory responsibility vested in the Board to protect the public health, it need only make a showing that the concentrations of particulate matter will tend to inflict injury.
The reason for the Division’s view is apparent from a review of the record. No expert testimony demonstrates any causal connection between the emissions from Duke City’s wigwam burner and actual injury to health. Moreover, other than the New Guinea study, see Duke City I, 95 N.M. at 406, 622 P.2d 709, which one physician describes as neither definitive nor conclusive, the parties have introduced no scientific papers which determine the correlation between wood smoke and health. Dr. Jonathan Samet, board-certified in internal medicine and pulmonary disease, testified that scientists have gathered little data for the purpose of analyzing health effects from wood smoke, since the problem “only recently surfaced because of increased residential woodburning.” In addition, Duke City’s burner has not been in existence long enough to determine actual adverse health effects, if any.
In order to determine the degree to which the parties must show that emissions will be injurious to health or safety, we look to the Environmental Improvement Act and the Air Quality Control Act. The purpose of the Environmental Improvement Act is “ * * * to ensure an environment that in the greatest possible measure: will confer optimum health, safety, comfort and economic and social well-being on its inhabitants; [and] will protect this generation as well as those yet unborn from health threats posed by the environment * * Section 74-1-2. In addressing a similar problem with respect to the promulgation of a primary air quality standard for lead, the D.C.. Circuit in Lead Industries Ass’n v. Environmental Protection Agency, 647 F.2d 1130 (D.C.Cir.1980), said that one purpose of the Clean Air Act is “ ‘[t]o emphasize the preventive or precautionary nature of the act, i.e., to assure that regulatory action can effectively prevent harm before it occurs * * *.' ” Id. at 1152. In speaking of the task the administrator has of making decisions about exactly what harms exist, the Court in Lead Industries said, “[M]aking these decisions is complicated by the absence of any clear thresholds above which therfc are adverse effects and below which there are none.” Id. at 1152. That statement applies to the case before us.
The Board in this case had to make a decision in an area where scientific knowledge has not to date provided an exact answer. To hold that a party must establish actual injury to health or safety would, under the circumstances of this case, not only thwart the purposes of the act in question, but would relegate the agencies charged with its enforcement to reacting to catastrophies after the fact. We do not believe the legislature intended this result. Therefore, we hold that while the Board may not rely on guesswork in making a finding as to injury to health or safety, neither must it base its finding on proof of actual harm. If we applied the stricter standard suggested by Duke City, then it would not have met its burden of making a prima facie showing; Duke City has not itself shown to a degree of medical probability that emissions from its wood burner will not be injurious to health or safety. The resolution of this case will not depend upon the proof or lack of proof of actual injury to health or safety.
III. Duke City’s Prima Facie Showing.
In Duke City I we held that Duke City had the initial burden of proving that granting the variance will not result in a condition injurious to health or safety. Because of the difficulty of that task, we did not require Duke City to offer plenary proof. We said: “ ‘It is enough that [Duke City] introduces such evidence as, in the absence of counter testimony, will afford reasonable ground for presuming the allegation is true * * V ” 95 N.M. at 405, 622 P.2d 709.
Duke City asserts that this Court must reverse as a matter of law, because after it made a prima facie showing, the Division failed to produce any “competent evidence of probable health or safety effects.” Duke City contends that it met its burden and that substantial evidence does not support the Board’s ruling. Duke City makes the following argument: The emissions from its burner do not violate the National Ambient Air Quality Standards (NAAQS), and even if they did, that would not in and of itself establish the existence of a condition injurious to health or safety. No medical evidence demonstrates injury to health or safety; the only published medical study indicates no detrimental effects.
The Board and Division counter by claiming that Duke City’s air-quality expert was not qualified, that his testimony was “riddled with uncertainties and inconsistencies,” and that the results of his tests were unreliable. As to the medical evidence introduced by Duke City, the Board and Division argue it was not sufficient to prove the negative. Finally, the Board and Division say Duke City failed to make a prima facie showing as to “safety,” arguing that it differs from “health.”
Both sides seem to agree that the NAAQS as set by the United States Environmental Protection Agency (EPA) provide relevant standards for determining adverse effects on public health. We note that AQCR 402A, promulgated by the State of New Mexico, differs from the NAAQS. AQCR 402A concerns measurement at the source, i.e., the plume of a particular wood waste burner, and operates in terms of opacity, while NAAQS concerns measurement of particulate matter from many sources in the ambient air and operates in terms of weight of the particulate in a cubic meter of air. Admittedly, Duke City cannot comply with AQCR 402A, hence, its application for a variance. Section 74-2-8 provides relief by permitting variances as long as emissions are not injurious to health or safety. Since NAAQS concerns public health, the parties and the Board have used these standards to determine if Duke City’s violation of AQCR 402A can be tolerated under § 74-2-8.
The NAAQS for particulates is:
75 microgrms per cubic meter—annual geometric mean.
260 micrograms per cubic meter—maximum 24-hour concentration not to be exceeded more than once per year.
40 C.F.R. § 50.6 (1982).
On the basis of his modeling study, Duke City’s expert, Loren Crow, testified to the improbability that the maximum concentration of particulate in the vicinity of the burner would ever exceed the national standard for particulates (NAAQS). Mr. Crow also measured actual ambient air concentrations in the vicinity of Duke City’s burner. According to Mr. Crow, at no time did a sample show a concentration exceeding the 260 micrograms—24-hour concentration.
Even in the absence of medical proof, we are satisfied from a review of the record that Duke City made a prima facie showing of no injury to health or safety, at least to the extent required by Duke City I. While not strong, the proof offered by Duke City suffices to shift the burden of going forward to the Division. We now turn to the question of whether the Division carried its burden of rebutting that presumption.
IV. The Division’s Rebuttal.
As we stated at the outset, in accordance with § 74-2-9(C), we must consider only evidence supporting the Board’s finding in the light most favorable to support the finding. We disregard all other evidence. United Veterans Org. v. New Mexico Prop. App. Dept., 84 N.M. at 118, 500 P.2d 199.
The question before us is: Did the Division overcome the presumption raised by-Duke City that granting a variance to Duke City would not be injurious to health or safety?
A. Medical proof.
We have previously said that no medical proof from either side proves or disproves actual harm to health.
We hold that substantial evidence does not support the Board’s finding that medical proof established actual injury to health or safety. Substantial evidence means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. Toltec Intern., Inc. v. Village of Ruidoso, 95 N.M. 82, 619 P.2d 186 (1980). While medical testimony indicated that particulate matter from wood-burning can exacerbate health problems such as asthma, emphysema and chronic bronchitis, the Division offered no direct medical evidence that this will occur as a result of emissions from Duke City’s burner. The medical evidence standing alone does not support the Board’s conclusion that granting the variance will create a condition injurious to health or safety.
B. Citizen testimony.
In its statement of reasons, the Board found:
13. Unrebutted citizen testimony, and testimony and affidavits from local physicians, corroborates the expert testimony on the potential injurious health effects from continued Duke City emissions, and independently consitute [sic] reliable evidence of injury to health.
14. Citizen testimony also provided material evidence of adverse safety effects, including hazardous driving conditions resulting from Duke City’s woodsmoke.
This evidence included the statement of one physician, Dr. Hassemer, from the Española Valley who said that he and his children had suffered asthma attacks while driving through wood smoke emitted by Duke City and that one child had experienced coughing spells and eye irritation. Another resident said, “[S]chool children are forever with red eyes, irritated eyes, nose and even throat * * *.” A nurse at the Española Hospital stated that while attending patients in the obstetrical ward, she had experienced adverse effects from the woodsmoke.
Duke City challenges the Board’s reliance on this evidence on the grounds that: (1) not a single citizen was sworn; (2) no cross-examination was permitted; (3) information with respect to the injurious-to-health-or-safety issue is peculiarly within the knowledge of medical experts; and (4) as to Dr. Hassemer, there was no basis given for his opinions.
In order to accommodate the citizens who wished to express their views, the Board held a special evening session. By one witness’s count, 350 citizens attended the session.
From a review of the transcript it is clear that only a small percentage of those who appeared chose to testify. From reading the statements of those who did speak, it is equally clear that many approached the hearing with diffidence. Undoubtedly in an effort to encourage their input and allow as many as possible the opportunity to speak, the chairman announced at the beginning of the session that he would not swear any of the people unless their testimony became highly technical. No one voiced any objections regarding this procedure.
Under NMSA 1978, Evid.R. 603, every witness, before testifying, must declare by oath or affirmation that he will do so truthfully. The parties have not cited us to any authority in New Mexico exempting witnesses at an administrative hearing from that rule, and we have found none. Nevertheless, Duke City waived any error by failing to object at the time. Dettore v. Brighton Township, 91 Mich.App. 526, 284 N.W.2d 148 (1979).
At the end of the afternoon session and before adjourning for the evening session when citizens would be heard, the chairman told counsel:
Those people have been waiting for this hearing for a long time, they’ve expressed a sincere interest in it, and I want you to take into consideration that these people are not represented by counsel, and any badgering of those witnesses .or any undue pressure on those witnesses by counsel from either side, is— the Chairman is going to kind of try to jump in on their side, should that happen. (Emphasis added.)
Duke City claims that by this remark the chairman discouraged cross-examination. We disagree. What he discouraged was “badgering” or “undue pressure” of witnesses by “either side.” Given the circumstances we do not think that this comment amounted to anything more than an admonition to avoid objectionable questioning— something that would clearly come within the chair’s authority to control if it occurred.
Moreover, the chairman advised the audience more than once that the attorneys might cross-examine them. At the beginning of the session for citizen input, he told the audience, “I possibly, or one of the other attorneys, after you have spoken, may ask you questions on your opinions or your feelings, but we wanted each one of you to feel free to express those opinions to us * * He also informed the people that in addition to the attorneys, he would permit two non-lawyers to ask questions. At the conclusion of Dr. Hassemer’s testimony the chairman asked, “Any questions from anybody?”
We are satisfied that no due process rights were violated and that Duke City’s decision not to cross-examine stemmed from tactical judgment, not inhibition.
Next, Duke City argues that the citizens’ statements constituted nothing more than “lay opinions” and could have no probative value regarding the issue of injury to health or safety. While it is true that lay witnesses cannot express opinions as to matters which require expertise, a lay witness may give opinions that are rationally based on the perception of the witness. NMSA 1978, Evid.R. 701.
Further, the Board permitted lay witnesses to describe symptoms they had experienced even though the cause might have been beyond the scope of their knowledge. No one would question the ability of an injured person to describe the pain he or she experienced after an accident. Similarly, witnesses in this proceeding could describe their symptoms at times when Duke City burned wood waste. This type of evidence differs from the conclusory opinions rejected in Duke City I, 95 N.M. at 405, 622 P.2d 709.
Finally, Duke City claims that neither the petition signed by the local physicians, nor the statement of Dr. Hassemer was substantial evidence. We agree as to the petition. Duke City, however, made no objection. In order to preserve error, a party must object at the hearing; it cannot raise the issue for the first time on appeal. See NMSA 1978, Evid.R. 103. As to Dr. Hassemer, Duke City says no basis for his opinions existed. Much of what Dr. Hassemer said came from source material and was general rather than specific. He quoted from texts and resolutions passed by the State and local medical societies regarding air pollution. Duke City did not object and did not cross-examine the witness, although the chairman invited it. We hold that Duke City waived any error by not objecting. The admissibility of evidence is generally a matter of discretion with the trial court, and we will not reverse absent a showing of abuse of discretion. State v. Bell, 90 N.M. 134, 560 P.2d 925 (1977). This rule applies to administrative hearings as well. Having reviewed the record, we are persuaded that the Board did not abuse its discretion in admitting the citizen’s statements and these statements supported the Board’s findings.
C. NAAQS.
This brings us then to the question of whether exceedance of the NAAQS warrants denial of the variance. The Board found that violation of this standard per se creates a condition injurious to health. It also found that “ambient air particulate concentrations resulting from Duke City’s wood smoke emissions can from time to time reasonably be expected to equal or exceed the range of 500 micrograms per cubic meter on a 24 hour average.” This would exceed the 260 factor in the NAAQS by 240.
The parties differ sharply as to the effect of the NAAQS. Duke City argues that any exceedance of the standard, without more, fails to establish injury to health or safety. It contends that since the NAAQS includes a margin of safety, the standards are set at levels lower than those at which detectable health effects will occur. Thus, says Duke City, there can be excursions above the standard without causing injury. The Board and Division respond by saying that the margin of safety protects against effects not yet uncovered by research, and therefore, any exceedance cannot be tolerated.
We do not need to decide if a violation of the NAAQS constitutes per se a condition injurious to health or safety. Both parties have used this standard as a guideline—they differ only as to interpretation and application. When we speak of a “condition injurious to health or safety,” we do not necessarily refer to actual harm, only a condition that tends to cause harm to health or safety. The Air Quality Control Act and the Federal Clean Air Act are designed to prevent harm before it occurs.
With reference to the Division’s modeling results, Dr. Samet, testifying for the Division, stated that exceedance of the NAAQS reduces the margin of safety and that one can expect some adverse health effects as a result, such as exacerbation of chronic lung disease, including chronic bronchitis, emphysema and asthma. Although Dr. Samet did not quantify at what levels such exacerbation might occur, his testimony permits an inference to be drawn that subtle adverse effects can arise from excursions.
Although the evidence offered by the Division is by no means conclusive, we believe, given the lack of scientific information and considering the stated purpose of the Act, that substantial evidence exists to uphold the finding, if the underlying testimony as to exceedance withstands judicial review.
Duke City launches an attack on the data used by the Division and the Sierra Club in their models to predict hypothetical concentrations of pollutants. Modeling is an acceptable method to predict concentrations. See, e.g., PPG Industries, Inc. v. Costle, 630 F.2d 462 (6th Cir. 1980). Duke City contends that the Division erred in compiling its factual data and in making its assumptions, so that the Board acted arbitrarily and capriciously in relying on their model. As noted in Ethyl Corp. v. Environmental Protection Agcy., 541 F.2d 1 (D.C.Cir.1976), “We must look at the decision not as the chemist, biologist or statistician that we are qualified neither by training nor experience to be, but as a reviewing court exercising our narrowly defined duty of holding agencies to certain minimal standards of rationality.” In order to consider model results, courts must determine whether a rational basis exists for their reliability. There must be rational connections between the factual inputs, the modeling assumptions, the modeling results and the conclusions drawn therefrom. Sierra Club v. Costle, 657 F.2d 298 (D.C.Cir.1981). Duke City’s argument that the data input of the model was faulty asks us to weigh the evidence, in short, decide whose expert is right. When Duke City asked EID’s expert where he got his fuel feed rate information, he replied that it came from observations made by his personnel. Other than asking whether the information had been verified, Duke City asked no further questions pertaining to the matter. Based on our review, we hold the Division met its burden of showing that a rational basis existed for the reliability of the model results.
V. CONCLUSION
In summary, we hold that while there was insufficient medical evidence standing alone to establish actual injury to health or safety from concentrations of wood smoke, a rational basis did exist to support the Board’s finding; this basis appears both in the citizen statements and in the evidence of exceedance above the NAAQS. Because of the lack of scientific information as to levels beyond which injury to health will occur, the NAAQS provides the best standard available. Exceedance of that standard, as shown here, justifies denial of a variance. We have not held, and the evidence will not justify holding, that the emissions from Duke City’s wigwam burner have caused or even within a degree of reasonable medical probability will cause injury to health. We hold only that the evidence and the inferences to be drawn therefrom warrant a denial, because subtle effects resulting from these concentrations may tend to cause exacerbations of respiratory conditions. We recognize that this standard may not be adequate to establish a causal relationship in a personal injury or workmen’s compensation action but are of the opinion that it comports with the purpose of the Environmental Improvement Act and the Air Quality Control Act, as well as the Federal Clean Air Act.
Based on a review of the record, we affirm the Board’s denial of the variance.
IT IS SO ORDERED.
HENDLEY and LOPEZ, JJ., concur.