United States Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 January 2, 1985 MEMORANDUM SUBJECT: Regional Implementation of Modeling Guidance FROM: Joseph A. Tikvart, Chief Source Receptor Analysis Branch, MDAD (MD-14) TO: Regional Modeling Contact, Regions I-X Attached for your use is information on the implementation of modeling guidance. Attachment 1 is an excerpt of a memorandum from J. Wilburn to D. Tyler (dated November 13, 1984) which identifies several issues. Attachment 2 provides our response to these issues. It is our intent that the response merely reiterate the way in which we understand modeling guidance to be routinely implemented by all Regional Offices. However, having formalized that understanding, we believe that its circulation is desirable. If you have any questions, please call me. Attachments cc: Chief, Air Programs Branch, Regions I-X B. Turner D. Wilson bcc: J. Dicke W. Keith F. Schiermeier Attachment 1 (Excerpt of Memorandum from J. Wilburn to D. Tyler, Dated November 13, 1984) As discussed in this memo, we are quite concerned as to our credibility regarding the development and approval of SIP revisions and bubbles which consider complicated and involved modeling. While our Armco experience may be viewed by some as atypical, we feel that the problem is real enough to the point that we request guidance on the following three questions: 1. When do changes in EPA modeling procedures become official Agency policy? Do such forms as informal modeling protocols and consensus opinions developed at meteorologist meetings and workshops constitute official Agency policy? If so, how is management at the regional division and branch level informed of those decisions (i.e., are such decisions communicated by policy memorandum or must regional management be dependent upon regional participants at such meetings and workshops to accurately convey OAQPS's policy decision)? 2. How do changes in Agency modeling policy affect in progress modelinganalyses? Do policy changes in modeling procedures invalidate modeling protocols which accurately reflected modeling policy at the initiation of ongoing modeling analyses? If so, we would appreciate copies of all policy memorandum which communicated such policies. 3. Will it be necessary in order for Armco's bubble application to be concurred with by OAQPS, for Region IV to require Armco to submit a fourth revision to their modeling procedures which would provide an analysis of the 46 days with more than 6 hours of calm which have thus far been deleted for the submittal pursuant to the original protocol? If so, we would like an explanation of the rationale for this requirement in light of our discussion in this memo. Attachment 2 United States Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 December 24, 1984 MEMORANDUM SUBJECT: Regional Implementation of Modeling Guidance FROM: Richard G. Rhoads, Director /s/ Monitoring and Data Analysis Division TO: James T. Wilburn, Chief Air Management Branch, Region IV The purpose of this memorandum is to provide answers to the three questions posed in your November 13, 1984, memorandum to Darryl Tyler. I am responding because, although your memorandum expressed general concern about broad policy issues and the adequacy of staff communications regarding the Armco Alternative Emission Control Plan, your three specific questions were restricted to dispersion modeling issues. Regarding your first question: Changes in EPA modeling procedures become official Agency guidance when (1) they are published as regulations or guidelines, (2) they are formally transmitted as guidance to Regional Office managers, (3) they are formally transmitted to Regional Modeling Contracts as the result of a Regional consensus on technical issues, or (4) they are a result of decisions by the Model Clearinghouse that effectively set a national precedent. In the last case, such issues and decisions are routinely forwarded to all of the Regional Modeling Contacts. In order for this system to work, the Regional Modeling Contacts must be actively involved in all Regional modeling issues and they must be consulted on modeling guidance as necessary by other Regional personnel. Regarding your second question: The time at which changes in modeling guidance affect on-going modeling analyses is a function of the type of agreement under which those analyses are being conducted. On-going analyses should normally be "grandfathered" if (1) there is a written protocol with a legal or regulatory basis (such as the Lovett Power Plant) or (2) the analysis is complete and regulatory action is imminent or underway. If the analysis is based on a less formal agreement and is underway, the Regional Office should inform the source operators of the change and determine whether the change can be implemented without serious disruption to the analysis. If for some reason any previous analysis must be redone, then it should be redone in accordance with current modeling guidance. In any event, consequences of failing to implement current guidance should be discussed with the OAQPS staff (Helms/Tikvart) to ensure that inappropriate commitments are not made by the Regional Office. Regarding your third question: As previously discussed with your staff, the recent Armco modeling analysis is technically inadequate and not approvable so long as the approximately 46 days with calms are ignored. At the time the original protocol was developed, the deletion of calms was common practice because we had no consensus on technically valid procedures for addressing calms. However, (largely due to the assistance of RO IV staff in developing a technical solution to the calms issue) this practice was discontinued by consensus of the Regional Modeling Contacts who recommended immediate implementation of the new procedures (see Joe Tikvart's June 13, 1983, memo to Regional Modeling Contacts). The subsequent Armco analysis which ignored calms was, therefore, deficient since there is no rationale for "grandfathering" an analysis which was initiated after the new calms guidance was disseminated. This issue is no longer an issue since Armco has already submitted a reanalysis that addresses the calms issue.