OREGON ENVIRONMENTAL COUNCIL; Citizens for the Safe Control of the Gypsy Moth; Elaine Olsen and Glen Olsen, Plaintiffs,
and
Friends of the Earth; National Coalition Against the Misuse of Pesticides, Plaintiff/Intervenors,
v.
Leonard KUNZMAN, Director, State of Oregon, Department of Agriculture, State of Oregon, Department of Agriculture; United States Department of Agriculture; John R. Block, Secretary, United States Department of Agriculture; et al., Defendants.
and
Oregonians for Food and Shelter, Inc., Defendant/Intervenors.
Civ. No. 82-504-RE.
United States District Court, D. Oregon.
May 6, 1986.*633 Larry N. Sokol, Mary Gray Holt, Jolles, Sokol & Bernstein, P.C., Portland, Or., Michael D. Axline, John Bonine, Ralph Bradley, Bradley & Gordon, Eugene, Or., for plaintiffs and plaintiff/intervenors.
Charles H. Turner, U.S. Atty., Thomas C. Lee, Asst. U.S. Atty., Portland, Or., Dorothy R. Burakreis, R.W. Rodrigues, Elizabeth Ann Peterson, U.S. Dept. of Justice, Washington, D.C., John DiLorenzo, Jr., Brendan Stocklin-Enright, DiLorenzo & Dietz, Portland, Or., for defendants and defendant/intervenors.
*634 REDDEN, District Judge:
On April 26, 1985, 614 F. Supp. 657 (D.C. Or.1985), I issued an Opinion in which I found the worst case analysis contained in the Final Environmental Impact Statement for the Eradication and Suppression of Gypsy Moths (FEIS) to be in violation of 40 C.F.R. § 1502.8. I therefore issued an injunction prohibiting the use of carbaryl, diflubensuron, acephate, and trichlorfon, effective immediately in Oregon and on January 1, 1986, on a nationwide basis. Defendants now move to lift the injunction on the grounds that they have produced an addendum to the FEIS which brings the entire impact statement into compliance with 40 C.F.R. § 1502.8 and all other regulations. Plaintiffs vigorously oppose the motion on numerous grounds. After careful consideration of all arguments, and being mindful of the important issues and potential environmental risks, I grant defendants' motion and dissolve the injunction.
BACKGROUND
Following the issuance of my April 26, 1985 Opinion, defendants moved for reconsideration which I denied. In that Order I noted that defendants had three options: appeal, rewrite the worst case analysis or do nothing. Defendants decided to appeal and rewrite the worst case analysis. On February 18, 1986, defendants filed a "Final Addendum to the Final Environmental Impact Statement as supplemented-1985." That document contains three "appendices" to the FEIS. Appendix H is termed the "Plain Language Summary of the Health Risks." Appendix I is a verbal interpretation of the data presented in tabular form in Appendix F (the worst case analysis) and toxicity information presented during last year's trial. Appendix J contains public comments and agency responses on the plain language version of the worst case analysis.
STANDARD
My review of the FEIS and addendum is a limited one. I must determine whether the agency made a "good faith effort" to take environmental concerns into account. Silva v. Lynn, 482 F.2d 1282, 1284 (1st Cir.1973). To meet this requirement the impact statement must fully explain its inquiry, analysis and reasoning. Id.
The Second Circuit has explained the requirement as follows:
The ... [impact statement] must set forth sufficient information for the general public to make an informed evaluation ... and for the decisionmaker to `consider fully the environmental factors involved and to make a reasoned decision after balancing the risks of harm to the environment against the benefits to be derived from the proposed action' ... In so doing the EIS ... [gives] assurance that the stubborn problems or serious criticisms have not been swept under the rug.
Sierra Club v. United States Corps of Engineers, 701 F.2d 1011, 1029 (2nd Cir. 1983).
A court is limited to reviewing the agency's procedural duty and is not authorized to substitute its judgment for that of an agency concerning the wisdom or prudence of a proposed action. California v. Block, 690 F.2d 753, 761 (9th Cir.1982). See also Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 21, 96 S. Ct. 2718, 2730 n. 21, 49 L. Ed. 2d 576 (court cannot interject itself within the area of discretion of the executive as to the choice of the action to be taken).
My review is solely for the purpose of determining whether the Department of Agriculture complied with the National Environmental Policy Act, 42 U.S.C. § 4331 et seq., and its regulations. If I am satisfied that it has my review is at an end. Id.
DISCUSSION
The sole issue is whether the addendum presents the worst case analysis in a readable fashion and clearly and concisely identifies the risks associated with the four synthetic insecticides. Both sides have submitted *635 lengthy and detailed affidavits from experts explaining a myriad of factors to be considered when determining the readability of any given document. Plaintiffs experts advocate the use of statistical computer models which analyze the number of words per sentence as well as the length of the words used. Defendants would supplement this analysis with other subjective consideration. Plaintiffs counter this approach by arguing that it is too simplistic and that its application to this document is inappropriate.
Plaintiffs' experts are Dr. Mark Shinn and Dr. Gwyneth Britton. Dr. Shinn states that Appendices H and I are not readable by the general public. In reaching this conclusion Dr. Shinn subjected the addendum to a series of tests whereby words and phrases are randomly selected and fed into a microcomputer which has been programmed with several statistical models. In all, Dr. Shinn analyzed the addendum using five separate reading indices. These indices indicate that Appendix H reads at an eighth to eleventh grade level and that Appendix I is written at the college and graduate school level. Dr. Shinn states that it is widely accepted that the general public reads at the sixth grade level and thus the addendum is not readable and easily understood by its intended readers.
Dr. Britton states readability encompasses theoretical constructs drawn from developmental psychology, information processing, psycholinguistics and linguistics. While many outside factors influence the readability of a document, she states, reading formulas, such as those employed by Dr. Shinn, "are the best predictive device we have" to determine readability. Dr. Britton states that the addendum is not understandable by the general public. In assessing the document she employed several methods of processing words, sentences and the use of secondary communication tools (overlays, illustrations, tables, etc.). She found the readability level to be uneven throughout the addendum, thus making it more confusing. She advocates redoing the document with a targeted audience of an eighth grade reading level.
Defendants have offered an affidavit by Dr. John Campbell in support of their position that the addendum is readable. Defendants state that the addendum was submitted to Dr. Campbell on several occasions during the drafting for comments on how to make it more understandable. Dr. Campbell outlines the suggestions he made during that process and notes that each suggestion was adopted.
Specifically, Dr. Campbell suggested that "[b]ecause paragraphs contain explanations of key ideas, I suggested that the writers pay particular attention to sentence structure. The terms used to discuss the logic of a paragraph are unity (focus) and coherence (flow). That is, does a paragraph as a unit of logical thought focus on a discernible idea, and are the supporting sentences organized to help the reader follow the flow of information." Campbell affidavit, page 19. Dr. Campbell also suggested that the paragraphs contain topic sentences, which Dr. Britton finds "admirable." Britton affidavit, page 22.
Dr. Campbell explains that readability is determined at the "moment the reader's emotional, cognitive and linguistic backgrounds interact with each other, with the topic and with the proposed purposes for reading and with the author's choice of semantic and syntactic structures all within the setting" in which the reading is being done. Campbell affidavit, pages 5-6. Dr. Campbell then listed six reader variables to be considered. Dr. Britton states that "[m]y main disagreement with Dr. Campbell's definition of reading is that it is simplistic.... I agree that reading is an abstract process based partially on the interaction of the reader with the material. The reader variable [sic] listed are adequate, though I would add physiological state (i.e., hungry, frightened, etc.) as well as reading ability." Britton affidavit, pages 4-5.
Dr. Campbell criticizes Dr. Shinn's use of reading formulas as being too simplistic and not subjective enough. Dr. Britton challenges Dr. Campbell's methods as being *636 too subjective and simplistic. Dr. Shinn rebuffs Dr. Campbell's criticism because no question regarding the results of his "test" was raised. Dr. Campbell refutes Dr. Shinn's challenge on a variety of grounds. And so it goes, with each expert defending his or her methods and calling the other side's expert incorrect, ill-advised and against the generally accepted concepts held by their profession. Nevertheless, I have taken each expert's comments into consideration in my review of the addendum, paying attention to the red flags each side suggests either lends to or detracts from readability.
I have read the addendum in its entirety once before consulting the experts and once following the review of their affidavits. I understood it both times. I find that it adequately conveys the risks associated with the use of the four synthetic insecticides in an understandable and readable manner. The subject matter of the addendum is difficult in that it discusses chemicals with complex names which trip the tongue, and states hypothetical risks of worst case scenarios. However, those risks are communicated succinctly, clearly and understandably. While the experts disagree on whether the addendum should have been drafted at the sixth, eighth or even twelfth grade reading level, I note that Appendix H is written at eighth to eleventh grade reading levels. I find this level to be sufficient to communicate, to the intended reader, the risks associated with the use of the four chemical sprays.
Plaintiffs' argument that the government should have scrapped the FEIS and started anew, rather than issuing an addendum is appealing. However, simply because it might be better presented does not render the FEIS and its addendum legally inadequate. The risk information is presented in Appendix F, and explained concisely and succinctly in Appendix H. That is all that is required.
Plaintiffs also challenge the readability of the FEIS and the addendum on the grounds of inconsistencies in the information presented. Plaintiffs have identified what they allege are eleven separate areas of inconsistency. I will briefly address each of these.
Plaintiffs' first example is that the FEIS contains a comment which states that if even one baby is born with a defect, or one person in a million gets cancer, then the risk is too great. Plaintiffs state that this comment is in sharp contrast with the statements in the addendum that trichlorfon and carbaryl could cause birth defects. This comparison does not show contradictory statements.
First, the comment referred to is from a citizen and contained in the section of the FEIS where such comments are to be published. The statement was not made by the drafters. Also, the comments in the addendum that the two chemicals could cause birth defects is taken out of context. In both circumstances it was noted that no scientific data is available to show that these chemicals do cause birth defects. Information is available from in vitro studies that show some mutagenesis can occur at high doses. It was further noted that for purposes of the worst case analysis it would be assumed that the chemicals could cause birth defects and not that they did cause birth defects. It is inappropriate to use data from in vitro studies to state conclusively what will happen in an in vivo situation. An in vitro study is one in which human or mammalian cells are isolated, grown in a treatment culture and then examined for effects. In vitro studies do not consider the intact body's natural defense mechanism, or the interaction of other natural chemical substances in the body with the study chemical.
Plaintiffs next contend that the FEIS and addendum contain contradictions regarding the risks to sensitive populations. They state that comments contained in the FEIS and Appendix F indicate that all doses of the synthetic insecticides are safe, while comments in the addendum state risks are involved at certain exposures. This argument is without merit. The comments cited by plaintiffs are interpretations *637 of distinct sets of data. The FEIS explores other data available on the subjects and reaches the same conclusions stated in the addendum.
Moreover, the Appendix H comments referred to by plaintiffs are discussing hypothetical situations. Those passages use data from nonhuman studies and extrapolate it to guess or suppose what results could be manifested. One of the passages cited by plaintiffs indicates that carbaryl causes birth defects, and appears on page H-36. This comment refers to a study using dogs as the research animal. I note that the experts testified at last year's trial that the numbers from this study are not a reliable source when calculating human risks because dogs are ultra sensitive to carbaryl, compared to other mammals.
Plaintiffs' third contention is that the FEIS and Appendix H use a specific number for the calculation of the cancer risk associated with carbaryl, which number is stated as invalid in Appendix I. I do not agree with plaintiffs' analysis. The first two comments utilize a cancer risk potential of 0.057, which was obtained using the Eisenbrand linear model. Appendix I of the addendum is a verbal description of the calculations set out in the worst case analysis. The worst case analysis computes the cancer risk associated with N-nitroscarbaryl, a breakdown product of carbaryl, in several different ways. The comment in Appendix I to which plaintiffs cite is a description of a calculation of cancer risk that did not employ figures obtained from the Eisenbrand linear model. There is no contradiction here. The experts are all in agreement that the only way to calculate the cancer risk of carbaryl is through the use of a mathematical model. This is because carbaryl has never been shown to cause cancer, only its breakdown product N-nitrosocarbaryl. Also, it is uncertain whether N-nitrosocarbaryl is produced in the human body when carbaryl is ingested.
Plaintiffs' fourth contention is that the FEIS, Appendix F, Appendix H and Appendix I each contain a different cancer potency risk for carbaryl. This statement is incorrect. The figure at page 61 of the FEIS (3 in a billion) is the worst case cancer risk for all gypsy moth projects. This passage also refers the reader to page F-94-95 for further discussion. The figure mentioned at F-76 (4 in a hundred million) is for suppression projects only. The other figure is for eradication projects. The passage quoted from Appendix H is a parenthetical and not the figure actually used in the worst case analysis. The parenthetical states what the cancer risk would be if the dog study was used. As noted, the dog study is unreliable because of the hypersensitivity of dogs to carbaryl. The sentence preceding the parenthetical states: "[t]he weighted cancer risk for carbaryl from the formation of N-nitrosocarbaryl is 2.1 in a billion for eradication projects, 3.5 in a billion for suppression projects, and 5.5 in a billion for the combination of both." The final passage from Appendix I is explaining a particular set of calculations and is not the final worst case scenario. Plaintiffs have taken these figures out of context and fail to show contradictions.
Plaintiffs also contend that the FEIS and the worst case analysis contain different statements regarding no-observable-effect-levels (NOELS) and acceptable daily intake (ADI) levels. I do not agree. The statement in the FEIS that is cited by plaintiffs is an overall summary statement contained in the introduction to the FEIS. On the other hand, the comment cited by plaintiffs in the worst case analysis is an explanation of a specific table of numbers.
Plaintiffs' sixth contention is that the main body of the FEIS and the worst case analysis show different cancer risks for acephate. Plaintiffs claim that at page vi the FEIS states that the worst case risk of cancer from exposures to acephate can range up to two in a million and that Appendix F on page F-81 says that the risk is one in 100,000. That page notes that the worst case dietary risks are about 4½ in a million, and the worst case observer risks are about seven in a million. I see no contradictions.
*638 Plaintiffs' seventh criticism is also without merit. The claim is that the FEIS states one level of carbaryl exposure as safe and Appendix I states another. As noted, Appendix I is a verbal explanation of specific calculations set out in the worst case analysis. Accordingly, a recitation of a figure in Appendix I is not a comment on what levels of exposure of particular chemicals are acceptable. Rather, Appendix I is telling the reader what numbers were used in what calculations and why.
Plaintiffs' eighth claim of contradiction is that there are three different lowest NOELs for carbaryl set forth in the worst case analysis, Appendix H and Appendix I. Plaintiffs have taken these figures out of context. Each NOEL cited by plaintiffs is from a different source, and the reason all three figures are cited is to make available different studies and different recited NOELs.
Plaintiffs' ninth and tenth statements of contradiction concern residues and their persistence on vegetation. These two battles were fought at last year's trial and resolved for defendants. Essentially, plaintiffs want particular studies mentioned, discussed, considered and highlighted. When this is done and the studies are not used as a primary source in the FEIS, plaintiffs complain of conflicting information. When it is not done, plaintiffs complain that the government has failed to consider all available studies.
Plaintiffs' final contention is that the FEIS states that ponds and springs provide humans with a primary source of drinking water. The government has since withdrawn that statement. I see no contradiction here that should cause me to invalidate the FEIS.
Plaintiffs also argue that the addendum is unreadable because it uses words and phrases that could have been replaced by other more easily understood words and phrases. For example, in describing the possible side effects of the various chemicals, the addendum refers to "cholinesterase inhibition." Plaintiffs assert that the addendum should have used the term "nervous system damage" instead. Plaintiffs' argument is without merit. First, cholinesterase inhibition and nervous system damage are not synonymous. Second, when the term cholinesterase inhibition is first used in the addendum it is defined.
Plaintiffs also argue that the addendum should have used "can kill fetus" instead of "can cause maternal toxicity," "deadly dose" instead of "acute lethal dose" and "pregnancy" instead of "gestation." Plaintiffs cite other terms they deem inappropriate. I find that the use of these phrases does not render the addendum unreadable.
Plaintiffs argue that Appendix I reads at an even higher grade level than Appendix H, and is more complex and technical. Plaintiffs ask me to invalidate the FEIS on the grounds that Appendix I violates 40 C.F.R. § 1502.8. I decline to do so. Appendix I is a verbal description of the mathematical calculations included in Appendix F and is not a plain language interpretation of the worst case analysis. The government was not required to include it in this addendum. At trial Mr. Yost, one of the principal drafters of the regulations, testified that the readability requirement was designed to force EIS preparers to cull from the technical information the bottom line and state it succinctly. Mr. Yost states that the technical data used to reach the final decision need not be in the EIS so long as it is incorporated by reference. The readability requirement does not require that the calculations and equations be included in the EIS, much less verbally explained.
Plaintiffs make four additional arguments concerning the legality of the FEIS that do not go to the issue of readability. I find each of these arguments to be without merit but will briefly discuss them.
Plaintiffs argue that the Department of Agriculture has abdicated its responsibility to independently evaluate NOELs and ADIs and that this is in violation of NEPA. I do not agree. In doing the worst case analysis and assessing the risk associated with the use of these chemicals the department relied on data from studies done on *639 animals, specifically other mammals. Certain figures were used in the analysis. One figure is the NOEL. NOELs are the highest amount of chemical administered to the test animal at which no adverse effects were observed. To account for the difference between humans and other animals these NOELs are reduced, usually by a factor of 100 times. These reduced amounts are called acceptable daily intake levels (ADI). An ADI is that amount of a particular chemical which the EPA considers safe for human ingestion every day for seventy years with no ill effects.
Plaintiffs argue that in the worst case analysis the department used NOELs and ADIs established by the EPA and the World Health Organization (WHO). Plaintiffs claim this is the abdication of the department's responsibility.
In a previous Opinion I determined that the use of EPA and WHO generated NOELs and ADIs was reasonable. There was ample evidence and testimony presented to establish that the EPA is the agency charged with determining NOELs and ADIs for each of the chemicals. The USDA is not required to independently make such determinations. See Appendix J, pages 115-116, 119-120, response to comments number 7 and 15.
Plaintiffs argue that at several critical points the addendum relies on information not readily available to the public, and so violates 40 C.F.R. § 1502. Plaintiffs state that the addendum assumes that the risk of heritable mutations is no greater than the risk of cancer, and that this assumption was based on use of a linear model to estimate such risks. Plaintiffs state that when the government was asked to cite a credible source that quantitatively estimates the risk of such heritable mutations the government responded that it was generally accepted that no such models were available. Plaintiffs note that the government went on to state that a personal communication with Dr. Brusick indicated that the use of the linear cancer risk model would be appropriate to estimate risks of mutagenesis. It is this personal communication of which plaintiffs complain.
Dr. Brusick's comments and qualifications have been told to the public in the addendum and in the FEIS. The "personal communication" and the resources relied upon by Dr. Brusick are disclosed in the FEIS and the addendum. Appendix J at J-117, Appendix F at F-97-98.
Plaintiffs next argue that the worst case analysis lifetime exposure conclusions are based on data that is neither available nor accessible to the public. Plaintiffs argue that the government's response to their comment at J-125 of the addendum clearly shows that the government relied on "disparate, unidentified, and far flung documents" and is thus violative of 40 C.F.R. § 1502. This argument is without merit. The sources that the government relied on in reaching its estimates are listed in Appendix E of the FEIS, and the addendum at H-27. (The sources were also identified at trial by Dr. Schneeberger).
Plaintiffs' final argument is that the FEIS and the addendum relied on studies that were performed with unpublished Union Carbide methodology. This argument is also without merit. Plaintiffs admit that the methodology is disclosed, but not so the data used in developing that methodology. The information a scientist uses in developing a methodology comes from many sources, including, of course, his or her brain. A scientist does not just study data to develop a methodology, rather a methodology is developed to obtain data.
Plaintiffs next argue that the government has failed to respond to public comment, thereby violating 40 C.F.R. § 1503. Section 1503 requires the agency preparing an environmental impact statement to fully respond to public comments. This is to prevent cursory answers that do not address the public's true concern. Plaintiffs contend that defendants' responses to their comments were evasive and ignored the substance of the criticism offered.
*640 Plaintiffs note their comment that it was misleading for the government to state there was no evidence that any of the four chemicals have caused cancer in humans at any dose levels, when in fact no evidence of cancer in humans had been sought. Plaintiffs note that "the defendants responded that it is not misleading." Plaintiffs fail to state the entire response to the comment set forth at J-115. The full response is:
The sentence is not misleading; it is intended to provide some perspective on the use of worst case assumptions. The sentence would be misleading if it led the reader to think that the four chemicals cannot cause cancer. But the preceding sentence clearly indicates that we assumed that the chemicals could cause cancer. Adding the sentence that you suggest would reduce the readability level by inserting a double negative, and it would introduce an unverifiable assertion (it is impossible to prove that the four chemicals have not caused cancer).
Plaintiffs next contend that their comment 6 drew an incorrect response. Comment 6 concerned the use of NOELs and LOAELs in the worst case analysis. The government's response was that the reason NOELs and LOAELs were used was because of my prior ruling that the use of these figures was reasonable. Plaintiffs argue that because my Opinion approved the use of such figures generally, their comments concerning specific NOELs was not properly answered. I do not find this reasoning persuasive.
Plaintiffs also complain that when they asked the addendum drafters why they relied on the EPA ADI for carbaryl, the drafters responded that they did so because the EPA is the regulatory agency charged with establishing those values. This response is inadequate, plaintiffs argue, because it does not tell them why the drafters ignored the other studies which had established different ADIs for carbaryl. The response fully informs plaintiffs why the drafters used the ADI they did: because the EPA is charged with establishing uniform values for these types of measurements. It is also explained in Appendix H why the ADI which plaintiffs insist should have been used was not used (it was the ADI calculated using the data from the dog study). Moreover, Appendix H in a footnote at H-14, calculates the worst case risk if this ADI were used.
Plaintiffs contend that the comment that the worst case analysis and the addendum did not consider sources other than dietary for ingestion of N-nitrosocarbaryl when there are nitrogen oxides present in the atmosphere which could arguably combine with carbaryl and form N-nitrosocarbaryl. This comment was fully responded to at J-125. That response indicates that the ingestion of air borne N-nitrosocarbaryl was not considered because it has never been shown that carbaryl can combine with nitrogen oxides in atmospheric conditions to form N-nitrosocarbaryl. The only conditions under which that chemical reaction can take place is at 47°C and in an acidic environment. Plaintiffs' comment was fully answered.
Plaintiffs also contend that they advised the drafters in writing that the drafters estimation of residues on vegetation after 20 days was incorrect, and that the drafters failed to respond. This contention is without merit. See page J-126, response to comment 36.
Plaintiffs' final argument is that they commented that the averaging of N-nitrosocarbaryl cancer potency factors to estimate cancer risks is not proper in a worst case analysis. Plaintiffs state that the highest cancer potency should be used in such an analysis. Plaintiffs argue that defendants answer (leading authorities in risk assessment analysis employ an average of numbers in reaching their determination) is incorrect. I do not agree. The government responded fully to plaintiffs' comment.
CONCLUSION
Defendants' motion to dissolve the nationwide injunction is granted. Appendix H, which is a plain language rewrite of the worst case analysis, is readable and understandable *641 by its intended audience. The FEIS as supplemented meets the requirements of NEPA and the regulations governing its implication.