# Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/fr%3A2014-22447 ## Record - **Collection:** Federal Register - **Document type:** Proposed Rule - **Published:** September 29, 2014 - **Citation:** 79 FR 58434 ## Text DEPARTMENT OF HEALTH AND HUMAN SERVICES Food and Drug Administration 21 CFR Part 112 [Docket No. FDA-2011-N-0921] RIN 0910-AG35 Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption AGENCY: Food and Drug Administration, HHS. ACTION: Proposed rule; supplemental notice of proposed rulemaking. SUMMARY: The Food and Drug Administration (FDA or we) is proposing to amend certain specific provisions of the proposed rule, “Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption.” We are taking this action because the extensive information received in public comments has led to significant changes in our current thinking on certain key provisions of the proposed rule. We are reopening the comment period only with respect to the specific issues identified in this document. DATES: Submit either electronic or written comments on the proposed rule by December 15, 2014. Submit comments on information collection issues under the Paperwork Reduction Act of 1995 by December 15, 2014 (see the “Paperwork Reduction Act of 1995” section of this document). ADDRESSES: You may submit comments by any of the following methods. Electronic Submissions Submit electronic comments in the following way: • Federal eRulemaking Portal: http://www.regulations.gov. Follow the instructions for submitting comments. Written Submissions Submit written submissions in the following ways: • Mail/Hand delivery/Courier (for paper or CD-ROM submissions): Division of Dockets Management (HFA-305), Food and Drug Administration, 5630 Fishers Lane, Rm. 1061, Rockville, MD 20852. Instructions: All submissions received must include the Docket No. (FDA-2011-N-0921) for this rulemaking. All comments received may be posted without change to http://www.regulations.gov, including any personal information provided. For additional information on submitting comments, see the “Comments” heading of the SUPPLEMENTARY INFORMATION section of this document. Docket: For access to the docket to read background documents or comments received, go to http://www.regulations.gov and insert the docket number(s), found in brackets in the heading of this document, into the “Search” box and follow the prompts and/or go to the Division of Dockets Management, 5630 Fishers Lane, Rm. 1061, Rockville, MD 20852. FOR FURTHER INFORMATION CONTACT: Samir Assar, Center for Food Safety and Applied Nutrition (HFS-317), Food and Drug Administration, 5100 Paint Branch Pkwy., College Park, MD 20740, 240-402-1636. SUPPLEMENTARY INFORMATION: Table of Contents Executive Summary Purpose of the Supplemental Notice of Proposed Rulemaking Summary of the Major Provisions of the Supplemental Notice of Proposed Rulemaking Costs and Benefits I. Background II. Amendments to the Previously Published Proposed Rule A. Proposed Subpart A—General Provisions B. Proposed Subpart E—Standards Directed to Agricultural Water C. Proposed Subpart F—Standards Directed to Biological Soil Amendments of Animal Origin and Human Waste D. Proposed Subpart I—Standards Directed to Domesticated and Wild Animals E. Proposed Subpart R—Withdrawal of Qualified Exemption III. Preliminary Regulatory Impact Analysis A. Overview B. Regulatory Flexibility Analysis C. Unfunded Mandates D. Small Business Regulatory Enforcement Fairness Act IV. Paperwork Reduction Act of 1995 V. Analysis of Environmental Impact VI. Comments VII. References Executive Summary Purpose of the Supplemental Notice of Proposed Rulemaking To minimize the risk of serious adverse health consequences or death from consumption of contaminated produce, FDA published the proposed rule entitled, “Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption,” which would establish science-based minimum standards for the safe growing, harvesting, packing, and holding of produce, meaning fruits and vegetables grown for human consumption (78 FR 3504, January 16, 2013). FDA proposed these standards to implement section 105 of the FDA Food Safety Modernization Act (FSMA) (Pub. L. 111-353). The comment period for the proposed rule closed on November 22, 2013. Taking into account information we heard at public meetings, and based on a preliminary review of written comments submitted to the docket, currently available information, and our subsequent analysis of the proposed provisions in light of this information, we are proposing certain new provisions and certain amendments to our previously proposed provisions. Summary of the Major Provisions of the Supplemental Notice of Proposed Rulemaking We are reopening the comment period to seek public comment on the specific issues and amended and new proposed provisions that are discussed in this document, which include the following: (1) Proposed amendments to paragraph (a) of proposed 21 CFR 112.4 to exclude from coverage of the Produce Safety proposed rule those farms or farm mixed-type facilities with an average annual monetary value of produce (as “produce” is defined in § 112.3(c)) sold during the previous 3-year period of $25,000 or less (on a rolling basis); and corresponding revisions to the definitions of “very small business” and “small business” in proposed § 112.3(b) to apply the monetary value thresholds based on sales of produce; (2) proposed amendments to the definitions of “farm” in proposed § 112.3(c) responding to comments about overlap between the produce and preventive control rules, such that establishments that pack or hold produce that is grown or harvested on another farm would now be subject to the produce safety standards of proposed part 112 regardless of whether or not that farm is under the same ownership; and corresponding revisions to the definitions of “covered activity,” “harvesting,” “holding,” and “packing” in proposed § 112.3(c); (3) proposed amendments to § 112.44(c) to update the microbial quality standard for water that is used during growing of produce (other than sprouts) using a direct application method in a way that is consistent with the U.S. Environmental Protection Agency's (EPA) current recreational water standard, i.e., a geometric mean of samples not to exceed 126 colony forming units (CFU) of generic Escherichia. coli (E. coli) per 100 milliliters (mL) of water and (when applicable) a statistical threshold value of samples not to exceed 410 CFU of generic E. coli per 100 mL of water; (4) new proposed provisions within § 112.44(c) to incorporate additional flexibility and provide means to achieve the proposed microbial quality standard for agricultural water used for direct application during growing, i.e., by either applying a time interval (in days) between last irrigation and harvest using a microbial die-off rate of 0.5 log per day (proposed § 112.44(c)(1)); and/or applying a time interval (in days) between harvest and end of storage (including during activities such as commercial washing) using appropriate microbial die-off or removal rates, provided there is adequate supporting scientific data and information (proposed § 112.44(c)(2)); in addition, a new proposed provision to provide for an alternative microbial die-off rate between last irrigation and harvest in accordance with § 112.12; (5) proposed amendments to § 112.45(b) and new proposed provisions §§ 112.45(c) and (d) to provide tiered-approaches for specific testing frequency requirements to test untreated surface water as well as untreated ground water, which would enable testing at a reduced frequency than that proposed in the previously published proposed rule; (6) a new proposed provision § 112.45(e) to provide that a farm may meet the requirements related to agricultural water testing using the farm's own test results or data collected by a third party or parties, provided the water source(s) sampled by the third party or parties adequately represents the farm's agricultural water source(s) and all other applicable requirements are met; (7) proposed removal of the 9-month minimum application interval for use of raw manure in proposed § 112.56(a)(1)(i). FDA defers its decision on an appropriate time interval until FDA pursues certain actions, including a robust research agenda, risk assessment, and efforts to support compost infrastructure development, in concert with the U.S. Department of Agriculture (USDA) and other stakeholders. At this time, we do not intend to take exception to the continuation of adherence to the National Organic Program (NOP) standard; (8) proposed amendments to § 112.56(a)(4)(i)) to establish that if the biological soil amendment of animal origin is treated by a composting process and is applied in a manner that minimizes the potential for contact with covered produce during and after application, then the minimum application interval ( i.e., time between application and harvest) is 0 days; (9) new proposed provision § 112.84 to explicitly state that part 112 would not authorize or require covered farms to take actions that would constitute the “taking” of threatened or endangered species in violation of the Endangered Species Act, or require covered farms to take measures to exclude animals from outdoor growing areas, or destroy animal habitat or otherwise clear farm borders around outdoor growing areas or drainages; (10) new proposed provision § 112.201(b)(1) to establish that, before FDA issues an order to withdraw a qualified exemption, FDA may consider one or more other actions to protect the public health and prevent or mitigate a foodborne illness outbreak, including a warning letter, recall, administrative detention, refusal of food offered for import, seizure, and injunction; (11) new proposed provisions §§ 112.201(b)(2) and 112.201(b)(3) to establish that, before FDA issues an order to withdraw a qualified exemption, FDA must notify the farm of circumstances that may lead FDA to withdraw the exemption, and provide an opportunity for the farm to respond to FDA's notification; and that FDA must consider actions taken by the farm to address the circumstances that may lead FDA to withdraw the exemption; and (12) new proposed provision § 112.213 to list the circumstances under which FDA would reinstate a farm's qualified exemption that is withdrawn. We are seeking comment on the issues discussed in this document by December 15, 2014. The previously published proposed rule (78 FR 3504; January 16, 2013) and the proposed amendments and new provisions published in this document, taken together, constitute the entirety of the proposed rule on “Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption.” At this time, we are not seeking comment on any other provisions of the previously published proposed rule that are not identified for public comment in this document. We will complete our review of public comments received thus far, and take into account comments received in response to this document, in issuing a final rule. Costs and Benefits We performed additional analyses to examine the impacts of the amended and new proposed provisions described in this document. We estimate the costs of the proposed rule as currently amended to be $386.23 million annually for domestic farms, $143.39 million annually for foreign farms covered by the rule (for a grand total of $529.62 million annually), resulting in $400.37 million annually in estimated potential net benefits. Summary of Costs and Benefits of the Proposed Rule as Currently Amended Prevented foodborne illnesses (in millions) Total benefits (in millions) Total domestic costs (in millions) Total foreign costs (in millions) Total costs (domestic + foreign) Net benefits (in millions) 1.57 $930.00 $386.23 $143.39 $529.62 $400.37 Compared to the previously published proposed rule, in total, this represents a cost savings of $73.33 ($459.56 − $386.23) million for domestic produce farms, and a decrease in overall net benefits of $7.19 ($400.37 − $407.56) million. I. Background To minimize the risk of serious adverse health consequences or death from consumption of contaminated produce, FDA published the proposed rule, “Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption” (hereafter referred to as “the Produce Safety proposed rule” or “the previously published proposed rule”), which would establish science-based minimum standards for the safe growing, harvesting, packing, and holding of produce, meaning fruits and vegetables grown for human consumption (78 FR 3504, January 16, 2013). We later issued a notice to correct technical errors and errors in reference numbers cited in the proposed rule (78 FR 17155, March 20, 2013). In the same issue of the Federal Register in which the Produce Safety proposed rule was published, FDA published another proposed rule entitled, “Current Good Manufacturing Practice and Hazard Analysis and Risk-Based Preventive Controls for Human Food” that would apply to human food and require domestic and foreign facilities that are required to register under the Federal Food, Drug, and Cosmetic Act (FD&C Act) to have written plans that identify hazards, specify the steps that will be put in place to minimize or prevent those hazards, monitor results, and act to correct problems that arise (hereafter referred to as “the Preventive Controls for Human Food proposed rule”) (78 FR 3646, January 16, 2013). These proposed rules help form the foundation of, and a central framework for, a new food safety system in the United States. We requested comments on the Produce Safety proposed rule by May 16, 2013. We extended the comment period for the proposed rule and its information collection provisions (which are subject to review by the Office of Management and Budget under the Paperwork Reduction Act of 1995 (44 U.S.C. 3501-3520) (78 FR 11611, February 19, 2013; 78 FR 24692, April 26, 2013; 78 FR 48637, August 9, 2013; 78 FR 69605, November 20, 2013). The comment period for the proposed rule closed on November 22, 2013. Since publication of the Produce Safety proposed rule in January 2013, we conducted numerous outreach activities. For example, we held three public meetings to solicit oral stakeholder and public comments on the proposed rule, inform the public about the rulemaking process (including how to submit comments, data, and other information to the rulemaking dockets), and respond to questions about the proposed rule (78 FR 6762, January 31, 2013, and 78 FR 10107, February 13, 2013). We also traveled across the country and around the world to discuss the Produce Safety proposed rule, as well as the other foundational FSMA proposed rules (Refs. 1, 2, and 3). II. Amendments to the Previously Published Proposed Rule In December 2013, FDA issued a public statement reiterating our goal of ensuring produce safety, and indicating that, based on the extensive input we have received from produce farmers, consumers, and others in the agricultural sector, significant changes will be needed in key provisions of the Produce Safety proposed rule, including those related to water quality standards and testing, standards for using raw manure and compost, certain provisions affecting mixed-use facilities, and procedures for withdrawing the qualified exemption for certain farms (Ref. 4). We also announced our intent to propose revised regulatory requirements and request comment on them, allowing the public the opportunity to provide input on our current thinking. In this document, FDA is providing our current thinking on certain issues discussed in the Produce Safety proposed rule that we previously published, including certain amended and new proposed provisions, for public comment. In addition, published elsewhere in this issue of the Federal Register , we are also providing our current thinking on certain issues discussed in the Preventive Controls for Human Food proposed rule that we previously published, and seeking public comment on those issues. To date, over 15,000 electronically submitted comments have been received in the docket in response to the previously published proposed rule. We are continuing to review all electronic and paper submissions of comments to the docket. Taking into account information received at public meetings, and based on a preliminary review of written comments submitted to the docket, currently available information, and our subsequent analysis of the proposed provisions in light of this information, we are reopening the comment period to seek public comment on certain specific issues described in this section. Importantly, the amended and new proposed provisions we have included in the regulatory text are based on a preliminary review of the comments. We will complete our review of comments previously submitted and consider the comments responsive to this document in developing the final rule. The previously published proposed rule and the proposed amendments and new provisions published in this document, taken together, constitute the entirety of the proposed rule on “Standards for the Growing, Harvesting, Packing, and Holding of Produce for Human Consumption.” Throughout this document, we discuss amendments and additions to the previously proposed part 112 and, in the codified section of this document, we list each of the amended and new proposed provisions of proposed part 112. For the convenience of readers and ease of reference, we prepared a separate document to identify the changes to the previously published codified provisions and provide the complete proposed part 112, as amended through this document (Ref. 5). A. Proposed Subpart A—General Provisions In the previously published proposed rule, under subpart A of proposed part 112, we proposed various provisions to establish the scope of, and definitions applicable to, the Produce Safety regulation, and to identify who would be subject to the requirements of part 112. Proposed subpart A also described the proposed modified requirements and procedures governing qualified exemptions from the regulations. We discussed each of the proposed provisions and explained our rationale (78 FR 3504 at 3534 through 3551). We are reopening the comment period to solicit public comment on our current thinking on two specific issues related to the general provisions in subpart A: (1) Farm sizes based on monetary value of total food sales to determine those farms that are not covered by the rule and those that would qualify for extended compliance periods and (2) definition of “farm”. We describe our current thinking on these two issues in this section. 1. Farms Sizes Based on Monetary Value of Total Food Sales In the previously published proposed rule, we proposed to apply the Produce Safety regulation only to farms and farm mixed-type facilities with an average annual monetary value of food (as defined under the FD&C Act and including seeds and beans used to grow sprouts) sold during the previous 3-year period of more than $25,000 on a rolling basis (proposed § 112.4). We explained that farms below this $25,000 limit collectively account for only 1.5 percent of covered produce acres, suggesting that they contribute little exposure to the overall produce consumption. Based on a tentative conclusion that such businesses do not contribute significantly to the produce market and, therefore, to the volume of production that could become contaminated, we tentatively concluded that imposing the proposed requirements of part 112 on these businesses is not warranted because it would have little measurable public health impact. We also noted that such farms are and would continue to be subject to the applicable provisions of the FD&C Act and applicable implementing regulations, irrespective of whether they are included within the scope of the Produce Safety proposed rule (78 FR 3504 at 3518 and 3549). In addition, we proposed to apply certain monetary value thresholds based on total food sales to define those very small and small businesses that would be eligible for our proposed extended time periods to comply with the Produce Safety regulation. In proposed § 112.3(b)(1), we proposed to define “very small business” to mean a business that is subject to proposed part 112 and for which, on a rolling basis, the average annual monetary value of food (as defined under the FD&C Act and including seeds and beans used to grow sprouts) sold during the previous 3-year period is no more than $250,000. In addition, under proposed § 112.3(b)(2), we proposed to define “small business” to mean a business that is subject to proposed part 112 and for which, on a rolling basis, the average annual monetary value of food (as defined under the FD&C Act and including seeds and beans used to grow sprouts) sold during the previous 3-year period is no more than $500,000, and which farm is not a “very small business.” a. Relevant comments. We received several comments opposing our tentative decision to identify farm sizes based on total food sales either for coverage by the rule or for extended compliance periods. Commenters recommended that farm sizes should be based on the sale of “covered produce” or “regulated” produce, rather than on the sale of all food. Some of these commenters noted that the proposed coverage of farms based on their total food sales would make it difficult for midsize farms to diversify their operations. Other commenters maintained that covering farms based on their total food sales would have an adverse impact on diversified farms that primarily raise food grains or dairy cattle (and produce dairy products) by forcing them to comply with produce safety standards. Some commenters that recommended identifying farm sizes (both those that would not be covered and those that would be covered and considered as “small businesses” or “very small businesses”) based on monetary value of sales of “covered produce” also recommended similarly applying the qualified exemptions provided under proposed § 112.5 to farms based on an average annual monetary value of $500,000 or less of sales of covered produce, rather than on sales of all food. b. FDA's consideration of comments. In response to comments received, we considered what, if any, revisions are needed to the proposed $25,000 limit as the threshold above which farms would be subject to the Produce Safety regulation. As noted in the previously published proposed rule, farms with an average annual monetary value of food sold of $25,000 or less collectively account for 1.5 percent of covered produce acres, suggesting that they contribute little exposure to the overall produce consumption. Applying the $25,000 limit to an average annual monetary value of produce (rather than food) sold would account for an estimated total of 4 percent of covered produce acres and about 3.1 percent of all produce acres in the United States. The amended proposal would remove farms with produce sales of $25,000 or less from coverage, resulting in removal of an additional 2.1 percent of produce acres from coverage (after removal of acres as a result of the provisions related to the qualified exemption, produce that is rarely consumed raw, and produce destined for commercial processing that eliminates pathogens of concern). Under this scenario, as with the previous proposed approach, such businesses would not contribute significantly to the volume of produce in the marketplace that could become contaminated and, therefore, would have little measurable public health impact. We believe that applying the $25,000 limit to produce sales would accommodate the concerns expressed by some comments without adversely affecting the level of public health protection, envisioned under our previous proposal. We also considered applying the $25,000 limit to average annual monetary value of “covered produce” sold, as requested by some commenters. This scenario presented a number of challenges, including the difficulty of determining the scope and public health impact of excluding farms based on the sales of covered produce, particularly considering the likely variability in produce commodities grown year to year; variability resulting from provisions under which certain commodities would not be considered “covered produce” (for example, produce that is rarely consumed raw); changes in the amount of produce that is used for personal consumption or for consumption on the farm or another farm under the same ownership; and whether and how to account for produce that would be eligible for exemption under certain conditions, which may be inherently variable based on market conditions (for example, produce that is destined for commercial processing). Given these confounding factors and based on available data, at this time, we are unable to determine the extent to which businesses with an average annual monetary value of “covered produce” sold of more than $25,000 would contribute to the overall produce market or the public health impact of not covering such businesses under part 112. In addition, the likely frequent change to a farm's covered or non-covered status may also be challenging for compliance and enforcement purposes. For these reasons, we are proposing to amend paragraph (a) of proposed § 112.4 to establish that if you are a farm or farm mixed-type facility with an average annual monetary value of produce (as “produce” is defined in § 112.3(c)) sold during the previous 3-year period of more than $25,000 (on a rolling basis), you are a “covered farm” subject to this part, and that if you are a “covered farm” subject to this part, you must comply with all applicable requirements of this part when you conduct a covered activity on “covered produce.” In addition, we are proposing corresponding revisions to the definitions of “very small business” and “small business” to apply the monetary thresholds consistently across three size-based categories of businesses that we proposed in the previously published proposed rule. As revised, a very small business defined under proposed § 112.3(b)(1) would mean a farm that is subject to part 112 and, on a rolling basis, the average annual monetary value of produce (as defined in proposed § 112.3(c)) sold during the previous 3-year period is no more than $250,000. As revised, a small business defined under proposed § 112.3(b)(2) would mean a farm that is subject to part 112 and, on a rolling basis, the average annual monetary value of produce (as defined in proposed § 112.3(c)) sold during the previous 3-year period is no more than $500,000; and the farm is not a very small business. Applying the monetary value limits for very small and small businesses to produce rather than to food, as previously proposed, would not alter the coverage of these businesses under the Produce Safety regulation, although we expect that a greater number of farms would likely fit within the revised definitions of very small business and small business and, therefore, qualify for the extended compliance periods we proposed for these businesses in the previously published proposed rule. See Table 1 for summary of these three proposed size-based categories, as revised. We seek comment on our current proposal to cover farms with an average annual monetary value of “produce” sold of more than $25,000, and the corresponding revisions to apply the relevant monetary thresholds to the sales of produce to define small businesses and very small businesses that would be subject to this regulation for the purpose of establishing extended compliance periods. We also seek comment on whether and how these monetary thresholds may be applied to covered produce only. Table 1—Summary of Proposed Qualifications [On a rolling basis, average annual monetary value of produce (as defined in proposed § 112.3(c)) sold during the previous 3-year period] Above $250,000 and no more than $500,000 Small Business. Above $25,000 and no more than $250,000 Very Small Business. $25,000 or less Not covered. We also considered applying the monetary value limit to covered produce sales, rather than to total food sales, in the criteria applicable to farms that would be eligible for a qualified exemption under proposed § 112.5. In the previously published proposed rule, we proposed that a farm would be eligible for a qualified exemption and associated modified requirements in a calendar year if: (1) During the previous 3-year period preceding the applicable calendar year, the average annual monetary value of the food (as defined in proposed § 112.3(c)) you sold directly to qualified end-users (as defined in proposed § 112.3(c)) during such period exceeded the average annual monetary value of the food you sold to all other buyers during that period; and (2) the average annual monetary value of all food (as defined in proposed § 112.3(c)) you sold during the 3-year period preceding the applicable calendar year was less than $500,000, adjusted for inflation (proposed § 112.5(a)). As explained in the proposed rule, proposed § 112.5(a) establishes the criteria for eligibility for a qualified exemption and associated special requirements based on average monetary value of all food sold and direct farm marketing, as mandated by section 419(f) of the FD&C Act (21 U.S.C 350h(f)). The criteria established in proposed § 112.5(a), including the requirement that “all food” be considered in calculating sales, are derived from section 419(f) of the FD&C Act. We, therefore, as a result of the statutory language, cannot apply the monetary value limit to covered produce sales, but instead must apply it to total or “all” food sales. Therefore, we are not able to make any change to the provision that the average annual monetary value of all food (as defined in proposed § 112.3(c)) sold during the 3-year period preceding the applicable calendar year must be less than $500,000, as proposed in § 112.5(a)(2)). 2. Definition of “Farm” (and “Covered Activity,” “Harvesting,” “Holding,” and “Packing”) In the previously published proposed rule, under subpart A of proposed part 112, we proposed definitions for various terms used in part 112. In proposed § 112.3(c), we proposed to define “farm” to mean to mean a facility in one general physical location devoted to the growing and harvesting of crops, the raising of animals (including seafood), or both. As proposed, the term ” farm” included: (1) Facilities that pack or hold food, provided that all food used in such activities is grown, raised, or consumed on that farm or another farm under the same ownership; and (2) facilities that manufacture/process food, provided that all food used in such activities is consumed on that farm or another farm under the same ownership. We also proposed definitions for “farm mixed-type facility” and related activities, such as “manufacturing/processing,” “packing,” and “holding.” In developing these definitions, we considered the interrelationship between farms and facilities, and articulated five organizing principles to explain the basis for the proposed definitions that would classify activities on-farm and off-farm for the purpose of the Produce Safety regulation. See the discussion of this issue in the previously published proposed rule (78 FR 3504 at 3539 through 3544). a. Relevant Comments. We received numerous comments regarding the proposed definition of a “farm,” including concerns related to packing or holding activities that routinely take place on a farm that commenters believed should be considered under the farm definition but would be instead covered under the proposed definition of a “mixed-type facility.” In particular, commenters noted that, as proposed, packing or holding of produce would be subject to either the Preventive Controls for Human Food regulation or the Produce Safety regulation, depending on whether or not the produce was grown on a farm under the same ownership. Commenters expressed various concerns with this proposed approach, including that: (1) This divergence in requirements does not have a public health basis given that the activities of packing or holding would present similar food safety risks regardless of the ownership of the farm on which the produce was grown; (2) subjecting a farm to the requirements of two different food safety regulations would be burdensome and is arbitrary; (3) it is common practice for a farm to buy and resell produce from other farms in order to fill out the necessary scale of supply (for example, when supplied to restaurants, retail establishments, or large wholesale markets), to pack produce for a neighbor who lacks a packing house, hold produce with a long shelf-life for a neighboring farm with limited storage space, or to pack or hold produce grown on farms of different ownership given costs associated with packing or holding activities; and (4) some farms sell their produce through “Community Supported Agriculture” arrangements and such deliveries often include produce grown by other farms not under the same ownership. We also received another comment that opposed broadening the proposed “farm” definition due to concerns that such changes could undermine the public health objectives of the rule. b. FDA's Consideration of Comments. We tentatively concur with commenters who stated that packing or holding of produce presents similar reasonably foreseeable hazards regardless of whether the produce is grown and harvested on farms under the same or different ownership, and that such hazards associated with packing or holding activities would best be addressed through the standards established under the Produce Safety regulation. In response to the comments described above and similar public comments received on the Preventive Controls for Human Food proposed rule, elsewhere in this issue of the Federal Register , we are issuing a notice to reopen the docket and seek public comment on certain specific issues related to that proposed rule (referred to as “amendments to the Preventive Controls for Human Food proposed rule”). In that document, we are proposing a revised definition of “farm” to mean an establishment under one ownership in one general physical location devoted to the growing and harvesting of crops, the raising of animals (including seafood), or both. As revised, the term “farm” would include establishments that, in addition to these activities: (1) Pack or hold raw agricultural commodities (RACs); (2) pack or hold processed food, provided that all processed food used in such activities is either consumed on that farm or another farm under the same ownership, or is processed food identified in subparagraph (3)(ii)(A) of the “farm” definition; and (3) manufacture/process food, provided that: (i) All food used in such activities is consumed on that farm or another farm under the same ownership; or (ii) Any manufacturing/processing of food that is not consumed on that farm or another farm under the same ownership consists only of: (A) Drying/dehydrating RACs to create a distinct commodity, and packaging and labeling such commodities, without additional manufacturing/processing; and (B) Packaging and labeling RACs, when these activities do not involve additional manufacturing/processing. Under this amended proposed definition of “farm,” farms that pack or hold produce RACs that are grown on a farm that is under a different ownership would no longer necessarily be “farm mixed-type facilities” subject to the requirements of the Preventive Controls for Human Food regulation. Rather, packing or holding others' produce RACs on a covered farm would now be subject to the Produce Safety standards of proposed part 112 (unless the produce is not covered by part 112, as described in proposed § 112.2). Similarly, we are proposing in that document to amend the definitions of “harvesting,” “holding,” and “packing,” consistent with this amendment to the farm definition and in response to other issues discussed in that document. We refer you to the discussion of this issue in section V of that document. Consistent with our proposed amendments to the definition of “farm” as it applies to proposed 21 CFR part 117, we are proposing to amend the definition of “farm” as it applies to proposed part 112 to include within that definition establishments that pack or hold RACs that are grown or raised on another farm, whether or not under the same ownership. In addition, we are proposing corresponding revisions to the proposed definitions of “covered activity,” “harvesting,” “holding,” and “packing” in proposed § 112.3(c) to remove the previous proposed restriction to encompass only RACs grown on farms under the same ownership. As revised, “covered activity,” “harvesting,” “holding,” and “packing” would encompass relevant activities regardless of the ownership of the farm where the RACs are grown. In the amendments to the Preventive Controls for Human Food proposed rule, we are also proposing certain other amendments to the definitions of “farm,” “holding,” and “packing,” taking into account comments received. For example, as amended, the proposed definition of “farm” also includes establishments that manufacture/process food by drying/dehydrating RACs to create a distinct commodity, and packaging and labeling such commodities, without additional manufacturing/processing. The amended proposed definition of “farm” also includes manufacturing/processing food by packaging and labeling RACs, when these activities do not involve additional manufacturing/processing. In addition, the amended proposed definition of farm would refer to “establishments” rather than to “facilities,” a term used in the previous proposed definition. In addition, as a conforming change relevant to this substitution, we are adding to the “farm” definition the criterion that the establishment is “under one ownership,” to retain that aspect of the current “farm” definition in the revised definition. As amended, the proposed definition of “holding” also includes activities performed incidental to storage of a food ( e.g., activities performed for the safe or effective storage of that food and activities performed as a practical necessity for the distribution of that food (such as blending of the same RAC and breaking down pallets)). Finally, as amended, the proposed definition of “packing” also includes activities performed incidental to packing a food ( e.g., activities performed for the safe or effective packing of that food (such as sorting, culling and grading)). We refer you to the discussion of these issues in section V of that document. Consistent with our proposed amendments to these definitions as they apply to proposed part 117, we are proposing to amend the definitions of “farm,” “holding,” and “packing” as they apply to proposed part 112. Taken together, we are proposing to amend the definition of “farm” in proposed § 112.3(c) to mean an establishment under one ownership in one general physical location devoted to the growing and harvesting of crops, the raising of animals (including seafood), or both. The term “farm” would include establishments that, in addition to these activities: (i) Pack or hold RACs; (ii) Pack or hold processed food, provided that all processed food used in such activities is either consumed on that farm or another farm under the same ownership, or is processed food identified in subparagraph (iii)(B)(1) of the “farm” definition; and (iii) Manufacture/process food, provided that: (A) All food used in such activities is consumed on that farm or another farm under the same ownership; or (B) Any manufacturing/processing of food that is not consumed on that farm or another farm under the same ownership consists only of: (1) Drying/dehydrating RACs to create a distinct commodity, and packaging and labeling such commodities, without additional manufacturing/processing; and (2) Packaging and labeling RACs, when these activities do not involve additional manufacturing/processing. As amended, “harvesting” would apply to farms and farm mixed-type facilities and means activities that are traditionally performed on farms for the purpose of removing RACs from the place they were grown or raised and preparing them for use as food. Harvesting is limited to activities performed on RACs on a farm. Harvesting does not include activities that transform an RAC, as defined in section 201(r) of the FD&C Act (21 U.S.C. 321(r)), into a processed food as defined in section 201(gg) of the FD&C Act. Gathering, washing, trimming of outer leaves of, removing stems and husks from, sifting, filtering, threshing, shelling, and cooling RACs grown on a farm are examples of harvesting. In addition, as amended, “holding” would mean storage of food and also includes activities performed incidental to storage of a food ( e.g., activities performed for the safe or effective storage of that food and activities performed as a practical necessity for the distribution of that food (such as blending of the same RACs and breaking down pallets)), but does not include activities that transform an RAC, as defined in section 201(r) of the FD&C Act, into a processed food as defined in section 201(gg) of the FD&C Act. Holding facilities could include warehouses, cold storage facilities, storage silos, grain elevators, and liquid storage tanks. Finally, as amended, “packing” would mean placing food into a container other than packaging the food and also includes activities performed incidental to packing a food ( e.g., activities performed for the safe or effective packing of that food (such as sorting, culling and grading)), but does not include activities that transform an RAC, as defined in section 201(r) of the FD&C Act, into a processed food as defined in section 201(gg) of the FD&C Act. (For reference, we previously proposed to define “packaging” (when used as a verb) to mean placing food into a container that directly contacts the food and that the consumer receives.) The defined term “covered activity,” which establishes what activities are subject to proposed part 112, is directly related to and affected by the scope of the definitions of “farm,” “harvesting,” “packing,” “holding,” and “manufacturing/processing.” We are proposing to amend the definition of “covered activity” to mean growing, harvesting, packing, or holding covered produce on a farm. “Covered activity” includes manufacturing/processing of covered produce on a farm, but only to the extent that such activities are performed on RACs and only to the extent that such activities are within the meaning of “farm” as defined in this chapter. This part does not apply to activities of a facility that are subject to 21 CFR Part 110 of this chapter. We are proposing these changes to the definition of “covered activity” to reflect the changes we are proposing to the definitions of “farm” and related terms. First, we have removed the limitation “provided that all covered produce used in covered packing or holding activities is grown, raised, or consumed on that farm or another farm under the same ownership” to reflect our proposed expansion of the farm definition to include packing and holding of others' produce RACs. Second, because we are proposing to add some additional, limited types of “manufacturing/processing” to the definition of “farm,” (and to reclassify some activities from “packing” to “manufacturing/processing”) those activities should be subject to proposed part 112 when they are performed on a covered farm on covered produce. For example, because the proposed definitions would now provide that packaging RACs would be manufacturing/processing (rather than “packing”), and would be within the farm definition if the packaging does not include additional manufacturing/processing, that activity should be covered by proposed part 112 when performed on a covered farm on covered produce. For example, a covered farm placing strawberries in a plastic “clamshell” package should be considered a “covered activity”. We seek comment on the amended definition of “farm,” and the corresponding changes to the definitions of “harvesting,” “holding,” “packing,” and “covered activity.” In addition, we seek comment on whether the phrase “in one general physical location” should be included in the farm definition in the final rule. We are aware that numerous produce farms own and grow crops in non-contiguous parcels of land in various geographical locations, such as in multiple States or even in more than one country. If finalized as proposed, how should we interpret “in one general physical location” for the purposes of enforcing this regulation? For example, farms that are in separate geographical locations, although under the same ownership, could be considered as different “farms” under this proposed definition and, therefore, such businesses might qualify for extended compliance periods that we proposed for “small business” and “very small business” farms. In addition, we seek comment on whether to include in the final rule a requirement that a farm supplying produce to another farm that will pack or hold that produce should provide to the farm that receives the produce its name, complete business address, and description of the produce in any individual shipment. Under these circumstances, is it appropriate to also require the farm that receives the shipment to maintain such record of information and, if so, for what specified period of time? Farms that pack or hold produce that is grown and harvested on farms under a different ownership and that are currently subject to the recordkeeping requirements of Subpart J of 21 CFR Part 1 may no longer be required to establish or maintain such records, if they fit within the amended proposed “farm” definition. Information about where the produce was grown or harvested may be important to trace contaminated product during an illness outbreak or other adverse event related to that produce and, therefore, we seek comment on whether we should require such farms to continue to be subject to recordkeeping requirements. Finally, we seek comment on whether on-farm packinghouses under cooperative ownership by multiple growers should be considered under the same ownership as any or all of the growers' farms, for the purposes of this regulation. 3. Summary of FDA's Revisions and Request for Comment We are proposing to: (1) Revise paragraph (a) of proposed § 112.4 to so that farms or farm mixed-type facilities with an average annual monetary value of produce (as “produce” is defined in § 112.3(c)) sold during the previous 3-year period of $25,000 or less (on a rolling basis) would not be covered by the Produce Safety regulation; and to make corresponding revisions to the definitions of “very small business” and “small business” in proposed § 112.3(b) to apply the monetary value thresholds based on sales of produce; (2) revise the definition of “farm” in proposed § 112.3(c) such that establishments that pack or hold produce RACs that are grown or harvested on another farm would now be subject to the Produce Safety standards of proposed part 112 regardless of whether or not that farm is under the same ownership; and corresponding revisions to the definitions of “covered activity,” “harvesting,” “holding,” and “packing” in proposed § 112.3(c); and (3) revise the definitions of “farm,” “holding,” and “packing” as they apply to proposed part 112, consistent with the proposed amendments as these terms apply to proposed part 117. We seek comment on our amended proposed provisions, including our current proposal not to cover farms with an average annual monetary value of “produce” sold of $25,000 or less and whether (and, if so, how), as an alternative, we should apply this monetary threshold to covered produce only. We also seek comment on the amended proposed definitions of “farm,” “harvesting,” “packing,” “holding,” and “covered activity,” and whether the phrase “in one general physical location” should be included in the farm definition in the final rule. In addition, we seek comment on whether, in instances where a farm supplies its produce to another farm to pack, hold, or store the produce, the farms involved should be subject to a requirement to establish and maintain a record of such produce shipment for tracking purposes in the event of an illness outbreak. We also seek comment on whether on-farm packinghouses under cooperative ownership by multiple growers should be considered under the same ownership as any or all of the growers' farms for the purposes of this regulation. B. Proposed Subpart E—Standards Directed to Agricultural Water Under subpart E of proposed part 112, we proposed science-based minimum standards directed to agricultural water. Specifically, we proposed various measures regarding agricultural water sources and distribution systems (proposed §§ 112.41 and 112.42); requirements for treating agricultural water (proposed § 112.43); requirements for testing agricultural water (proposed § 112.44) and at certain specified frequencies (proposed § 112.45); requirements for water used in harvesting, packing, and holding activities (proposed § 112.46); and certain record-keeping requirements (proposed § 112.50). We discussed each of the proposed provisions and explained our rationale (78 FR 3504 at 3559-3573). We are reopening the comment period to solicit public comment on our current thinking on three specific issues related to the provisions for agricultural water: (1) Microbial quality standard for agricultural water used during growing activities for covered produce (other than sprouts) using a direct water application method; (2) frequency of testing agricultural water; and (3) use of third party agricultural water testing data. We describe our current thinking on these three issues in this section. 1. Microbial Quality Standard for Agricultural Water Used During Growing Activities for Covered Produce (Other Than Sprouts) Using a Direct Water Application Method In the previously published proposed rule, under proposed § 112.44(c), we proposed to require that when agricultural water is used during growing activities for covered produce (other than sprouts) using a direct water application method, you must test the quality of water in accordance with one of the appropriate analytical methods in subpart N. We also proposed that if you find that there is more than 235 CFU (or most probable number (MPN), as appropriate) generic E. coli per 100 mL for any single sample or a rolling geometric mean (n = 5) of more than 126 CFU (or MPN, as appropriate) per 100 mL of water, you must immediately discontinue use of that source of agricultural water and/or its distribution system for the uses described in proposed § 112.44(c). Moreover, before you may use the water source and/or distribution system again for the uses described in proposed § 112.44(c), we proposed that you must either reinspect the entire agricultural water system under your control, identify any conditions that are reasonably likely to introduce known or reasonably foreseeable hazards into or onto covered produce or food-contact surfaces, make necessary changes, and retest the water to determine if your changes were effective; or treat the water in accordance with the requirements of proposed § 112.43. As explained in the proposed rule, our review of available scientific literature led us to tentatively conclude that the above described standards, which are consistent with the EPA recreational water standards, provide an appropriate basis to establish the microbial quality standard for agricultural water that is applied to produce using a direct application method. We explained our rationale and acknowledged the challenges related to identifying an appropriate microbial quality standard for such use of agricultural water where the water is intended to, or is likely to, contact covered produce or food-contact surfaces during use of the water. For example, we acknowledged that these EPA standards were developed from epidemiological studies that correlated the risk of gastrointestinal illness to exposure to marine and freshwater by swimmers rather than to consumption of produce. These epidemiological studies were performed in beach areas subject to point source fecal contamination rather than non-point sources ( e.g., birds, agricultural and livestock runoff), which may impact agricultural water. We also noted that risks of adverse health outcomes resulting from full-body contact in contaminated water may be different than risks associated with consuming produce irrigated with contaminated water, given the differences in the expected routes of infection and pathogen mortality rates in the different environments (bodies of water for the EPA recreational water standards; soil, plants, and produce for this proposed rule). We considered other options, including proposing a standard based on the EPA drinking water standard or proposing a second lower microbial quality criteria for water used in growing, but where the water used for direct application is not reasonably likely to contact the edible portion of the covered produce. However, we did not find sufficient scientific support for such options. Moreover, we noted that we are aware that some industry groups have adopted the generic E. coli component of the EPA recreational water standards in the absence of sufficient information to support a pathogen-based microbiological standard for water used in the production of produce (78 FR 3504 at 3563 and 3569). Therefore, we tentatively concluded that the above described generic E. coli criteria would serve to minimize risk of known or reasonably foreseeable hazards when used as a standard for agricultural water used on produce (other than sprouts) during growing in a direct water application method. We discussed each of the proposed provisions and explained our rationale (78 FR 3504 at 3563 and 3569). a. Relevant Comments. We received an extensive number of comments on this issue, and a majority of them either questioned the scientific rationale for the proposed microbial quality standard, emphasized the burden placed on growers due to the stringency of this standard, and/or urged us to consider other factors that would allow the safe use of agricultural water that does not meet the proposed microbial water quality standard in direct application during growing activities. Commenters identified various concerns with the proposed microbial quality standard for agricultural water used during growing activities for covered produce (other than sprouts) using a direct water application method, including the following: (1) The lack of adequate data to inform a complete and thorough understanding of produce contamination resulting from irrigation water. Some commenters noted that there are relatively few confirmed cases of irrigation water as a source of pathogens in any food borne illness outbreaks, while other commenters thought that the proposed microbial quality standard appeared to address risks that are unidentified and unsubstantiated, without sufficient or meaningful underlying scientific rationale; (2) concerns with using the water quality standards developed for recreational water to determine acceptable levels of indicator organisms in agricultural water. Commenters opposed using the EPA standards and argued that it is not scientifically sound to apply the recreational water standards that are developed based on epidemiological data to irrigation water. Commenters also noted other limitations to this approach, including that using a recreational standard for water quality does not take into account the rapid die-off rate that occurs post irrigation and prior to harvest; (3) the need for education, guidance, and training to ensure growers understand the elements embedded in the proposed requirement and know how to properly sample, test, and make the necessary calculations to then determine whether or not their water meets the proposed microbial quality standard. Commenters also recommended simplifying the standard to eliminate the requirement for a rolling geometric mean (n = 5) of no more than 126 CFU (or MPN, as appropriate) per 100 mL of water, such that the single sample limit would then be maintained as the requirement. One commenter suggested that this proposed requirement would create an opportunity for confusion and noncompliance due to miscalculation or misunderstanding of the complex equation; (4) concerns that the proposed water quality threshold is either more stringent than or differs from other national or international best practices, recommendations, or guidelines. Commenters noted that the FDA proposed standard is more stringent than the World Health Organization (WHO) thresholds and urged us to amend the provisions to be more in line with WHO quality thresholds. Other commenters recommended following the Codex Alimentarius Commission's global standard (1,000 E. coli CFU/mL), the more stringent Canadian standard (77 E. coli CFU/100mL), or other thresholds established in the European Union; (5) concern that current science is inadequate to justify a fixed, generally applicable test organism, quantitative microbial quality threshold, or testing requirements. For example, one commenter asserted that a different microbial standard should be considered for overhead irrigation water that is applied prior to fruit set or more than 14 days before harvest because, under field conditions, water that does not meet recreational water quality standards would be quite safe for such use. Another commenter cited the lack of adequate scientific information to develop a generally applicable microbial quality standard, and recommended that FDA employ the generic E. coli standard as a voluntary measure until such time that more scientific information is generated and FDA develops an appropriate standard. Still others urged us to delay the use of a quantitative standard to allow for new scientific information to evolve in the future that would enable identifying microbial quality thresholds specific to different regions and types of water; (6) concern with the use of generic E. coli as an indicator to test safety of agricultural water, including water used in direct application. One commenter suggested including E. coli O157:H7, non-O157 Shiga toxin E-coli, Listeria monocytogenes, and Salmonella spp. as pathogens to be tested in water quality tests. Another commenter noted that researchers have found that levels of E. coli present in water used for crop sprays do not represent the microbial load on the surface of tomatoes at harvest. This commenter also pointed out that tests conducted by a major U.S. grower have demonstrated that the generic E. coli standard can be exceeded without human pathogens being present, and it can be met when human pathogens are actually present in high quantities, thus, bringing into question the reliability of generic E. coli as an appropriate indicator. Another commenter urged FDA to provide for flexibility to allow alternative indicators of water safety. This commenter pointed out that several States have replaced water testing programs with a risk based computer modeling approach to address recreational water safety, and instead of using test results to determine if recreational water is safe, computer modeling programs that calculate the risks of a given source are designed to accurately predict when water will be outside acceptable ranges. The commenter recommended that the final rule should allow flexibility within the agricultural water section to allow this approach when an appropriate model has been designed; and (7) concern that, in identifying the microbial standard for direct application, FDA failed to consider certain significant factors that affect whether and how the microbial standard is applied to irrigation water used in different regions of the country and for different types of crops. For example, it was noted that farms in Maine use a wide variety of water sources, from city water to wells to open water sources. Even with open water sources there is a wide variety including rivers, ponds, streams and some water bodies affected by ocean tides, which require site-specific timing for water use. Another commenter stated that, in the Treasure Valley, irrigation systems mix clean water with runoff water, and such inter-mixing results in high counts of E. coli in irrigation water throughout large parts of the water distribution systems during the growing season. We also received a comment stating that surface water in some regions or watersheds may regularly fail the generic E. coli test, and that 30 percent of the samples of water collected at 22 surface water sites in the southeastern Vermont region in 2012 had generic E. coli levels that exceeded 235 CFU per 100 mL. The commenter further explained that, without a real scientific justification, the rule would remove an important source of agricultural irrigation to farmers in that region at critical periods throughout the growing season. Yet another commenter pointed out that, in eastern Oregon, growers downstream will inherently have higher microbiological contaminant loads than those upstream, due to runoff reuse systems and other water conservation measures, and as proposed, the Produce Safety regulation will undoubtedly injure downstream growers by preventing them from utilizing their water for the use stated on their water permit or certificate. Finally, we also received a comment that asserted that, in some parts of the western United States where farmers do not control the water, it would be extremely burdensome for FDA to require testing and mitigation for unidentified and unsubstantiated risks that may not, in fact, exist. We also received several other comments in relation to the proposed requirement for testing water used for direct application. A commenter pointed out that the scientifically observed rates of microbial decline reported by some authors are vastly greater than the rates assumed in FDA's assessment of risks. The commenter disagreed with FDA's proposed microbial quality standard, and argued that FDA has chosen to regulate all directly applied agricultural water over the entire production season even though its own analysis supports regulating agricultural water only within a short window of a few days before harvest, thereby substantially increasing the costs associated with water quality testing with little substantiated benefit. Another commenter urged FDA to explicitly permit growers to use water testing data compiled by other entities. According to the commenter, municipalities in New Hampshire routinely test E. coli levels for recreational purposes, and it would be unnecessary to require growers to test the same water source for the same pathogens separately. In contrast, some other comments generally agreed with the use of a numerical standard for testing water quality. These commenters suggested that a numerical standard is necessary, particularly where the effectiveness of individual control measures, such as to protect the source of agricultural water from contamination, are either not properly implemented or not fully known. In such cases, a numerical standard would serve as an objective tool to monitor the water quality on a specified schedule and trigger corrective actions, where necessary. b. FDA's Consideration of Comments. As explained in the previously published proposed rule, based on a qualitative assessment of risk, we identified agricultural water as one of the most likely sources of produce contamination. Our tentative conclusions included: (1) There is a significant likelihood that surface waters may contain human pathogens, and surface waters pose the highest potential for contamination and the greatest variability in quality of the agricultural water sources; (2) susceptibility to runoff significantly increases the variability of surface water quality; (3) water that is applied directly to the harvestable portion of the plant is more likely to contaminate produce than water applied by indirect methods that are not intended to, or not likely to, contact produce; (4) timing of water application in produce production before consumption is an important factor in determining likelihood of contamination; and (5) microbial quality of source waters, method of application, and timing of application are key determinants in assessing relative likelihood of contamination attributable to agricultural water use practices (78 FR 3504 at 3522, 3523). Consequently, our proposed standards for agricultural water including those for microbial quality of water and testing frequencies for ground water and surface water, address these potential contributing factors. We do not believe that FDA should reconsider the use of generic E. coli as an indicator to test safety of agricultural water, including water used in direct application. As discussed in the previously published proposed rule, we proposed to use generic E. coli as an indicator of fecal contamination. We acknowledge that the presence of generic E. coli will not always correlate to the presence of pathogens in water. However, the presence of fecal contamination, especially as indicated by high levels of generic E. coli, may increase the likelihood of pathogen contamination in water (Refs. 6, 7, and 8). Therefore, the intent is to manage the presence of fecal contamination as a proxy for potential pathogen contamination, similar to use of fecal contamination as an indicator for the quality of water at swimming beaches and waters for harvesting molluscan shellfish (Refs. 9 and 10). In addition, several commenters noted that generic E. coli is an appropriate organism to use to characterize water quality and agreed with our proposal to require such characterization; these commenters expressed that generic E. coli provides the best and most practicable quantitative criterion at this time. Further, testing for pathogens to determine the appropriateness of the water would be more costly than testing for generic E. coli because of the need to test for multiple pathogens. We also acknowledge the limitations of a general requirement for agricultural water for growing using direct application that is based on a single microbial indicator and associated quantitative microbial quality threshold, in that it may not adequately account for differences in risk associated with irrigation practices used for different commodities. Although we are proposing to retain a single microbial quality requirement that would apply to all agricultural water for growing using direct application, our proposed new provisions in §§ 112.44(c)(1) and 112.44(c)(2) provide for flexibility in order to address comments that requested us to account for the wide range of irrigation water sources, irrigation practices in different regions of the country, and different types of crops. We also tentatively determined that a quantitative microbial standard that is enforceable and facilitates necessary action by industry to ensure the safe use of water when used for direct application would be more appropriate than a qualitative water quality standard. Taking into account comments received, currently available information, and upon further analysis, we are proposing amendments to proposed §§ 112.44(c), 112.44(d), and 112.50(b) that, collectively, result in the following changes: (1) Update the quantitative microbial quality requirements in a way that is consistent with the 2012 recreational water quality criteria (RWQC); (2) provide an allowance for microbial die-off between irrigation and harvest using a specified microbial die-off rate; (3) provide an allowance for microbial reduction between irrigation and end of storage; and (4) allow the use of an alternative in lieu of our specified microbial die-off rate between irrigation and harvest. The scheme outlined above, each element of which is discussed in more detail in the sections immediately below, is consistent with the construct of the standard recommended by the WHO, although less restrictive than that standard. The WHO approach rests on a multistep process to achieve incremental microbial reductions to meet the overall necessary scheme, yielding a tolerable disease burden due to raw produce consumption that is no greater than that adopted for drinking water (non-detectable E. coli per 100 mL) (Refs. 11a and 11b). The initial step of the multibarrier process begins with wastewater treatment, which is followed by subsequent preventive measures to achieve the final health-based target of ≤10 −6 DALY (disability-adjusted life year) per person, per year. Two specific examples of the multi-barrier process discussed in the guidelines are water qualities of 10 4 or 10 3 CFU E. coli per 100 mL, post-wastewater treatment, for use on surface and root crops, respectively, followed by subsequent mitigation strategies (Ref. 11a). According to the WHO analysis, using water of this microbial quality is dependent upon a 2-log reduction due to die-off between last irrigation and consumption (includes die-off in the field and during distribution) and a 1-log reduction attributed to washing prior to consumption. The WHO analysis recognizes the variable nature of die-off values, ranging from 0.5 to 2.0 log per day. FDA's previously proposed standard of 235 CFU generic E. coli per 100 mL for any single sample (or a rolling geometric mean of no more than 126 CFU per 100 mL) defined a microbial level for agricultural water used during growing activities using a direct water application method that would minimize the risk of serious adverse health consequences or death throughout the diversity of agricultural conditions, in addition to which alternatives could be developed to provide for the reductions assumed in the WHO standard for die-off in the field and during distribution and from activities such as washing. In response to comments asking for consideration of die-off and greater flexibility and to align with international recommendations from WHO and also Codex, we are again proposing a generally applicable microbial level for all agricultural water, but now allowing a standard reduction due to die-off in the field before harvest and consideration of additional die-off from activities such as storage or commercial washing. As described in the sections immediately below, these reductions would provide additional means to achieve our proposed microbial quality standard for agricultural water used in a direct application method of a statistical threshold value (STV) of 410 or less CFU of generic E. coli per 100 mL of water or a geometric mean (GM) of 126 or less CFU of generic E. coli per 100 mL of water, where known microbial reduction occurs after application. We believe that this approach is strongly supported by comprehensive risk management frameworks and associated recommendations for managing health risks in recycled wastewater use in agriculture (Refs. 11a and 12). As will be discussed in detail in section II.B.2., we are also proposing certain amendments to proposed §§ 112.45 that, collectively, result in a proposed tiered approach to testing untreated surface water and untreated groundwater. The proposed approach would allow farms to make decisions about safe use of available water sources prior to the beginning of the next growing season; adjust testing frequencies dependent on long-term test results; and ultimately reduce the required frequency of testing. i. Updating the quantitative microbial quality requirements. We continue to find that the EPA generic E. coli criteria for recreational water quality provides a quantitative microbial standard that is generally applicable to minimize the risk of known or reasonably foreseeable hazards associated with the use of agricultural water on produce other than sprouts during growing in a direct water application method. Further, the EPA analysis supporting the RWQC, while not perfect for our purposes, was developed using the necessary scientific rigor and describes illness rates due to incidental ingestion that can be generalized across different bodies of water (Ref. 13). In addition, while commenters objected to the use of RWQC to establish microbial quality requirements for agricultural water for growing using direct application, there is no consensus among commenters as to other appropriate alternative criteria or methodology. A majority of the concerns with using the RWQC appeared to center around the need to account for circumstances that are unique to produce growing and irrigation, such as die-off after application, which are factors that would not have been accounted for in formulating water quality requirements for recreational water purposes. We acknowledge these shortcomings, but we also believe that our complete set of amendments to proposed § 112.44(c), including new provisions in paragraphs (c)(1) and (c)(2), address these concerns. Therefore, we continue to see the value in using the EPA RWQC as the starting point for a quantitative microbial water quality standard for water that is used for growing of produce (other than sprouts) in a direct application method in proposed § 112.44(c) (with additional provisions in proposed §§ 112.44(c)(1) and (c)(2), as explained in sections II.B.1.b.ii. and II.B.1.b.iii.). In the previously published proposed rule, we proposed to use the EPA recreational water criteria that were published in 1986 for this purpose. In November 2012, EPA recommended new RWQC to update their 1986 criteria (Ref. 14) (hereafter referred to as “the 2012 RWQC”). Unlike the previous criteria, the 2012 RWQC specify a STV in conjunction with a recommended GM to describe the magnitude of the relevant bacterial indicators. The STV approximates the 90th percentile of the water quality distribution and is intended to be a value that should not be exceeded by more than 10 percent of the samples taken. The 2012 RWQC recommend a culturable E. coli level of a GM of 126 CFU per 100 mL of water and an STV of 410 CFU per 100 mL of water. The 2012 RWQC are based on several recent health studies and use a broader definition of illness to recognize that symptoms may occur without a fever, including a number of stomach ailments. Among other evidence, EPA considered the latest research that demonstrates a link between fecal contamination in recreational waters and illness, and designed the criteria to protect primary contact recreation where immersion and ingestion are likely. We refer you to EPA's 2012 RWQC and accompanying documents for a full description of the new criteria and the underlying scientific rationale (ibid.). Consistent with this new analysis, we are proposing to amend the microbial water quality standard in § 112.44(c) to reflect E. coli levels that are consistent with the recommendations in both the GM and STV values specified in the 2012 RWQC. As amended, proposed § 112.44(c) would require you to develop and verify the water quality profile of the water source as described in § 112.45(b)(1), and using your water quality profile as described in § 112.45(b)(1), take certain actions if you find that (when applicable) the estimate of the STV of samples exceeds 410 CFU of generic E. coli per 100 mL of water, or if you find that the GM of samples exceeds 126 CFU of generic E. coli per 100 mL of water, in order for you to use this water for direct application during growing of covered produce (other than sprouts). As amended, proposed § 112.44(c) would no longer include a maximum threshold of E. coli in a single sample of 235 CFU per 100 mL. Rather, a STV of water quality distribution of 410 CFU per 100 mL would be used when there are sufficient numbers of samples to calculate it, in conjunction with the GM in all cases. This standard would be similar to the 2012 RWQC in that regard. Adoption of the STV, which approximates the 90th percentile of the water quality distribution, as a criterion acknowledges the inherent variability of E. coli measurements in water systems, while continuing to be sufficiently protective of public health. In addition, use of the STV does not establish a single value that, if exceeded, would require immediate corrective action. Instead, any value above 410 CFU per 100 mL may be acceptable, as long as those values (each corresponding to a water sample) do not result in a calculation of STV that exceeds 410 CFU per 100 mL. For example, a water source found to contain 2,100 CFU generic E. coli per 100 mL in one of 10 samples analyzed may be appropriate to use in direct application during growing, if the remaining 9 samples are such that the STV (based on all 10 samples) is 410 CFU or less of generic E. coli per 100 mL of water. We seek comments on the absence of such a maximum level of generic E. coli, particularly in light of evidence that suggests that pathogen levels can increase at higher levels of generic E. coli or other indicators (Refs. 6, 7, and 8). In providing comments, we ask that you take into account that pathogens can survive for months in the soil and in crop tissue if they permeate that tissue, that soil or fecal material on the surface of produce may permeate cut tissues and create conditions to enhance the probability of growth of pathogens and other microorganisms, and that colonization and biofilm development may result in conditions that are protective for pathogens (Refs. 15 and 16). Some public comments, too, recommended that we consider the WHO recommended levels of 1,000 CFU per 100 mL and 10,000 CFU per 100 mL for root crops and surface crops, respectively, as adequate maximum E. coli levels. Note, however, that the WHO values are better explained as illustrations of how specific health protection measures could be used together after treatment ( e.g., treatment, die-off, and washing or treatment and drip irrigation) to achieve the additional log reductions recommended for waste water reuse. As such, those values are not to be viewed as absolute end point or maximum permitted levels. Rather, under new proposed provisions §§ 112.44(c)(1) and 112.44(c)(1), we are proposing to provide for a WHO-type scheme that could be used to satisfy the proposed requirements for microbial quality of water. For example, under this proposed approach, there would be no maximum threshold for a baseline of generic E. coli above which the agricultural water would be precluded from use in direct application during growing such that you would not be able to apply an appropriate time interval between last irrigation and harvest or between harvest and end of storage. We seek comment on whether we should establish a maximum level of E. coli (GM and/or STV) above which the water should not be permitted for use in direct application (until specific followup actions are taken to ensure it meets the recommended microbial quality requirements) and, if so, what would be an appropriate maximum level. As amended, proposed § 112.44(c) would continue to include a GM value of no more than 126 CFU per 100 mL of water, which is intended to be used in conjunction with the proposed STV explained above, consistent with the 2012 RWQC. However, we are removing the previously proposed requirement for a “rolling geometric mean (n = 5)” based on the sampling criteria we proposed in amended § 112.45(b), which is discussed in section II.B.2.b. According to the 2012 RWQC, the waterbody GM should not be greater than 126 CFU per 100 mL during any 30-day interval, and there should not be greater than a 10 percent excursion frequency of 410 CFU per 100 mL based on the calculated STV during the same 30-day period (Ref. 14). We considered whether to apply the 30-day interval of the 2012 RWQC as a sampling frequency, and tentatively conclude that this criterion would be difficult to apply in the context of our proposed sampling scheme. Instead, we are proposing amendments to proposed § 112.45 (see section II.B.2.) that would establish specific sampling frequencies ranging from 2 years for baseline characterization of water quality to annual verification of water quality. We agree with comments that cited the need for education to ensure that growers and other relevant staff are appropriately informed and trained to properly test and perform the necessary calculations to determine how best to use their water, particularly when it does not meet the proposed microbial quality requirements. We have tentatively determined that both the GM and STV values (when there are sufficient samples to calculate STV), which reflect the central tendency ( i.e., the extent to which statistical values fall around a middle value) of the water and its variability, respectively, are necessary parameters to properly characterize the water. We expect to issue guidance document(s) to assist with education and training to help farmers understand and implement any final requirements in § 112.44(c). We seek comment on our proposed amendments, including our decision to retain general microbial quality requirements and update them consistent with the 2012 RWQC; the use of GM and STV values to establish general microbial quality requirements; and the absence of a maximum generic E. coli threshold. ii. Allowance for microbial die-off between irrigation and harvest. In the previously published proposed rule, we acknowledged that in specific circumstances an alternative standard ( e.g., a standard that applies a time between application and harvest in place of the proposed § 112.44(c) standard, but is specific to a specific commodity or commodity group and region) may be appropriate if the alternative standard is shown to provide the same level of public health protection as the standard in proposed § 112.44(c) and not to increase the likelihood that the covered produce will be adulterated. Accordingly, under proposed § 112.44(d), we provided for the use of alternatives to the requirements in proposed § 112.44(c). We also noted that we are working with stakeholders to facilitate research into application intervals that would be commodity- and region-specific, such that water not meeting the proposed § 112.44(c) standard could be used in a direct water application method for growing covered produce other than sprouts as long as it was applied before the start of the scientifically established application interval ( i.e., at a certain number of days before harvest or earlier) (78 FR 3504 at 3553). Comments, however, included concerns from growers that buyers would demand that the grower meet the standard established in the Produce Safety regulation rather than meet an alternative that had not been explicitly sanctioned by FDA. A number of commenters that opposed our previously proposed microbial quality requirements also cited the lack of allowance for microbial reduction due to natural die-off in the field after application and prior to harvesting of the crop. On further consideration of this issue and relevant available scientific information, we are proposing to add a new provision under proposed § 112.44(c) to explicitly provide for use of water that meets the proposed microbial quality standard after accounting for microbial die-off, if applicable to your crop and practices on your farm. We discuss new proposed provision § 112.44(c)(1) in this section. Proposed § 112.44(c)(1) would provide one option by which you would be able to achieve the microbial quality requirements for agricultural water specified in § 112.44(c). Under this option, you must apply a time interval (in days) between last irrigation and harvest using a microbial die-off rate of 0.5 log per day to achieve a (calculated) log reduction of your GM of generic E. coli level to 126 CFU or less per 100 mL and of your STV to 410 CFU or less per 100 mL of water. Examples of 0.5 log per day calculations follow this discussion. Based on a review of currently available scientific literature, we tentatively determined that it would be appropriate to provide an allowance for microbial die-off between last irrigation and harvest using a proposed die-off rate of 0.5 log per day (Ref. 17). Survival of pathogens and other microorganisms on produce commodities is dependent upon several environmental factors, including sunlight intensity, moisture level, temperature, pH, the presence of competitive microbes, and suitable plant substrate. Generally, pathogens and other microbes die-off or are inactivated relatively rapidly under hot, dry, and sunny conditions compared to inactivation rates observed under cloudy, cool, and wet conditions. The impact of these variables results in a range of microbial die-off rates of 0.5 to 2.0 log per day (Refs. 11a and 12). We have evaluated the relevant studies and acknowledge that die-off rates below 0.5 log per day have been reported in the literature for particular crop and pathogen types, but we conclude that a rate of 0.5 log per day provides a reasonable estimate of die-off under a broad range of variables to include pathogen characteristics, environmental conditions, crop type, and watering frequency. FDA is currently engaged in research activities in this area. In an effort to support scientific research in the area of agricultural water, one of FDA's Centers of Excellence, the Western Center for Food Safety at University of California, Davis, partnered with the Center for Produce Safety to provide seed money through a competitive grants program to fund produce safety projects focused on agricultural water issues that are topical and/or region specific. Research areas that have received funding through this process include transfer and survival of organisms on produce after exposure from contaminated surface irrigation water, application of biocide technology on manure-contaminated irrigation water, the potential role of overhead sprinkler irrigation systems in the contamination of produce, and the survival of pathogens during the growing, harvesting, and storage of dry bulb onions after exposure with contaminated water. We seek comment on the appropriateness of the proposed 0.5 log per day die-off rate. Note also that the proposed provisions in § 112.44(d) would allow you to establish and use an alternative microbial die-off rate between last irrigation and harvest (in lieu of the proposed rate of 0.5 log per day), provided you satisfy the requirements of proposed § 112.12. When applying a microbial die-off rate of 0.5 log per day, as proposed, the time interval ( i.e., number of days) you apply between last irrigation and harvest are the days necessary to achieve the reductions in both the GM and STV values of generic E. coli to levels at or below those expected on produce if it were irrigated with agricultural water that satisfied the microbial quality requirements proposed in § 112.44(c). We tentatively conclude that use of such a time interval would provide the same level of public health protection as the standard in proposed § 112.44(c) and not increase the likelihood that the covered produce will be adulterated. This provision assumes that, for any given crop, the microbial levels found on produce after accounting for die-off when it is irrigated with water under the provisions of § 112.44(c)(1) would be approximately equal to or below the levels found if the crop were, instead, irrigated with water of higher quality ( i.e., that met our proposed microbial quality criteria). Reductions to achieve both GM and, when applicable, STV criteria are necessary to ensure that risk thresholds determined in the 2012 RWQC are not exceeded. For example, if you determined (using the procedures described in proposed §§ 112.45(b) or 112.45(c), as applicable), that your agricultural water which is to be used for the purposes described in § 112.44(c) has generic E. coli levels with a GM value of 241 CFU per 100 mL and a STV value of 576 CFU per 100 mL, your water would not meet the microbial quality specified in § 112.44(c), in that your values exceed both the GM value of 126 CFU per 100 mL and STV value of 410 CFU or less per 100 mL. Under proposed § 112.44(c)(1), you would be able to use this water by applying a calculated time interval of 1 day between your last irrigation event (by direct application method) and harvest of the crop. Using a microbial reduction rate of 0.5 log per day, a 1-day time interval would be sufficient to meet the microbial quality requirements specified in § 112.44(c) because it would reduce your GM and STV values to 76 CFU per 100 mL and 182 CFU per 100 mL, respectively. As another example, if you determined that your agricultural water has generic E. coli levels with a GM value of 241 CFU per 100 mL and a STV value of 4,600 CFU per 100 mL, your water would not meet the microbial quality requirements specified in proposed § 112.44(c). Under proposed § 112.44(c)(1), you would be able to use this water by applying a calculated time interval of 3 days between your last irrigation event (by direct application method) and harvest of the crop. Using a microbial reduction rate of 0.5 log per day, 3 days between irrigation and harvest would be sufficient to achieve a 1.5 log total reduction and reduce your GM and STV to 8 CFU per 100 mL and 145 CFU per 100 mL, respectively. We agree with comments that cited the need for education to ensure growers understand the elements embedded in our proposed requirements for agricultural water during growing using direct application. Relevant staff would need to be appropriately trained to properly sample, test, and make the necessary calculations to determine how best to use their water. We expect to work with the Produce Safety Alliance, and will also plan to issue guidance document(s), as needed, to further clarify our provisions and assist with such education and training, if these proposed provisions in § 112.44(c) are finalized, as proposed. In addition, there are resources available that would enable simply entering sample data into a form and automatically deriving the GM and STV values and/or calculating the appropriate time interval between irrigation and harvest, such that a farmer would not need to perform the necessary calculations. We plan to identify and provide such resources, if this proposal is finalized. We seek comment on our proposed approach and tentative conclusions, including the appropriateness of permitting an adequate time interval between last irrigation and harvest as a means to achieve the specified microbial quality requirements, and the appropriateness of using a microbial reduction rate of 0.5 log per day. In addition, we seek comment on whether we should require farms to establish and maintain any documentation in relation to the option to apply an adequate time interval between last irrigation and harvest, as provided in proposed § 112.44(c)(1). For example, should we require that farms must keep records that identify the time interval applied, how the time interval is calculated, and/or the dates of last irrigation and harvest corresponding to that time interval? iii. Allowance for microbial reduction between harvest and end of storage. A number of comments that opposed our previously proposed microbial quality requirements also cited the lack of allowance for microbial reduction due to natural die-off during storage and/or due to pathogen removal during certain post-harvest activities, such as commercial washing, prior to consumption. On further consideration of these issues and relevant available scientific information, we are proposing to add another new provision under proposed § 112.44(c). We discuss the new proposed provision § 112.44(c)(2) in this section. Proposed § 112.44(c)(2) would provide a second option by which you would be able to achieve the microbial quality requirements specified in § 112.44(c). Under this option, you must apply a time interval (in days) between harvest and end of storage using an appropriate microbial die-off rate between harvest and end of storage and/or appropriate microbial removal rates during activities such as commercial washing to achieve a (calculated) log reduction of your GM of generic E. coli level to 126 CFU or less per 100 mL and (when applicable) of your STV to 410 CFU or less per 100 mL, provided you have adequate supporting scientific data and information. You may apply this time interval in addition to the time interval in accordance with 112.44(c)(1). This provision would allow you to apply appropriate microbial die-off or reduction rates post harvest ( i.e., between harvest and end of storage, and during activities such as commercial washing), provided you have adequate supporting scientific information. As discussed in the section immediately above, we expected that farms would consider such factors as microbial die-off or microbial reduction post irrigation and prior to consumption, as they are applicable to their commodity and/or practices on the farm, and apply appropriate scientifically-supported alternatives (such as time intervals) under the provisions we proposed in § 112.44(d). However, based on comments, we are proposing new provision § 112.44(c)(2) to incorporate additional flexibility into our agricultural water quality standards, and provide farms with yet another means by which to safely use agricultural water by achieving our proposed microbial quality requirements, without compromising the safety of produce that comes into contact with such water. As previously noted, the WHO study attributed a 1-log reduction in microbial load to washing (Ref. 11a). In addition, it is reasonable to expect some die-off during post-harvest storage, though the rate would be highly dependent upon the conditions of storage. Farms would be able to more narrowly define die-off rates associated with their specific production practices and apply a time interval (in days) between harvest and end of storage, calculated using microbial die-off rate(s) for the period between harvest and end of storage, including any microbial removal rate(s) as a result of commercial washing, as applicable to their commodity. Regardless of the microbial rates applied, the total log reduction necessary and the time interval required would need to be calculated based on a comparison of the GM and (when applicable) STV values of your agricultural water with the proposed microbial quality requirements (GM of 126 CFU or less per 100 mL and STV of 410 CFU or less per 100 mL) in § 112.44(c). At this time, we are not proposing to establish a specific microbial die-off rate(s) between harvest and end of storage or a specific microbial removal rate(s) during post-harvest activities such as commercial washing that can be broadly applied to calculate an adequate time interval between harvest and end of storage. We do not have sufficient information to support the derivation of an appropriate broadly applicable microbial reduction rate(s) between harvest and end of storage, or during activities such as commercial washing. However, under this option, you would be able to establish and apply an adequate time interval using a microbial die-off rate(s) that is relevant to your covered produce and dependent on practices and conditions on your farm, provided you have adequate scientific data or information to support your conclusions. As we noted in the previously published proposed rule, we are working with our stakeholders to facilitate research into application intervals that would be commodity- and region-specific, such that water not meeting the proposed § 112.44(c) standard could be used in a direct water application method for growing covered produce (other than sprouts) as long as it was applied before the start of the scientifically established application interval ( i.e., at a certain number of days before harvest or earlier). We will disseminate the results of these investigations, when available, and issue commodity- and region-specific guidance as appropriate, such that farmers would be able to consider our recommendations and apply the new scientific information to their current use of agricultural water, as appropriate. In addition, we are proposing to add a new provision, i.e., proposed § 112.50(b)(8), to require you to establish and keep records of such scientific data or information you rely on to support the microbial die-off or removal rate(s) that is used to determine the time interval (in days) between harvest and end of storage and/or other activities such as commercial washing, as applicable, used to achieve the calculated log reduction of generic E. coli in accordance with the provision in § 112.44(c)(2). This record-keeping requirement would enable us to verify the scientific basis for your time interval, should you choose to employ the approach permitted in § 112.44(c)(2). As in the case of alternatives permitted under § 112.12, we are not proposing to require farms to submit scientific data or information relied on to support the microbial die-off or removal rate applied in accordance with § 112.44(c)(2) to us for review or approval prior to marketing produce grown under those conditions. However, we would require that farms maintain a record of any such scientific data or information, including any analytical information, and make such data and information available to us to evaluate upon request. We seek comment on this proposed provision, including on whether there is a specific microbial die-off rate(s) or microbial removal rate(s) that we should establish within this provision. We also seek comment on whether and, if so, how we should introduce additional flexibility. iv. Provision for use of an alternative microbial die-off rate. As explained in section II.B.1.b., we are proposing to add a new provision § 112.44(c)(1) related to agricultural water used in a direct application method to permit the use of an adequate time interval between last irrigation and harvest, based on a microbial die-off rate of 0.5 log per day, to achieve water quality that meets the proposed microbial standard. We acknowledge that practices and conditions on a farm and circumstances unique to a specific commodity or types of commodities could result in higher die-off rates, especially under conditions of high ultraviolet radiation, high temperature exposures or low humidity, coupled with little precipitation. To account for such variability, we are proposing a new provision, i.e., proposed § 112.44(d)(2), to specify that you may establish and use an alternative microbial die-off rate (in lieu of the 0.5 log per day microbial rate that we proposed under § 112.44(c)(1)), to determine the time interval (in days) between last irrigation and harvest, provided you satisfy the requirements of § 112.12. Among other requirements, the use of an alternative microbial die-off rate would necessitate you to have adequate scientific data and information to support your conclusions. We refer to section V.B of the previously published proposed rule for a discussion of the requirements of § 112.12. Finally, as amended, proposed § 112.44(c) would continue to retain the previously proposed option to discontinue the use of water that does not meet the proposed microbial quality requirements and take corrective actions, prior to using that water for the same purposes. Proposed § 112.44(c)(3) would establish a third option, in lieu of following the procedures in §§ 112.44(c)(1) or 112.44(c)(2), where if water does not meet the proposed microbial quality requirements, you would immediately discontinue use of that source of agricultural water and/or its distribution system for the uses described in § 112.44(c). Before you may use the water source and/or distribution system again for those uses, you would be required to either reinspect the entire agricultural water system under your control, identify any conditions that are reasonably likely to introduce known or reasonably foreseeable hazards into or onto covered produce or food-contact surfaces, make necessary changes, and retest the water to determine if your changes were effective; or treat the water in accordance with the requirements of § 112.43. 2. Frequency of Testing Agricultural Water In the previously published proposed rule, under proposed § 112.45, we proposed to establish requirements related to frequency of testing agricultural water that is subject to the requirements of proposed § 112.44. Specifically, proposed § 112.45(a) would require that you test any agricultural water that is subject to the requirements of § 112.44 at the beginning of each growing season, and every 3 months thereafter during the growing season, except that there would be no requirement to test water that meets certain conditions specified in proposed § 112.45(a)(1) to (a)(3) (i.e., treated water and water from a public water system). As explained in the previously published proposed rule, water testing frequencies recommended by various industry documents vary widely, in part, because there is a lack of publicly available information pertaining to the quality of agricultural waters. Recommendations range from monthly testing to once each year, for sources with a history of compliance with commodity specific recommendations. Even for sources considered reliable ( e.g., well water), a 1-year period between testing may not minimize the risk of known or reasonably foreseeable hazards because microbiological water quality is often too variable for this frequency of testing to be protective ( e.g., effects of flooding, runoff). Alternatively, we tentatively concluded testing well water more frequently than every 3 months would not significantly improve the accuracy of your assessment of ground water quality and would therefore be unnecessary. We also considered proposing testing frequencies as a function of commodity, irrigation method ( e.g., furrow, seep, subsurface drip, foliar), and timing of application (days prior to harvest), and concluded that the most effective approach is to test at a frequency related to the reliability of the agricultural water sources. We requested comments on whether we should allow for adjustment of ground water testing frequencies dependent on historical test results, for example, testing ground water sources every 3 months for 1 year and yearly after that if the ground water consistently met the standard. We also requested public comments on any other alternative testing frequencies that can be supported by water quality data (78 FR 3504 at 3570). In addition, under proposed § 112.45(b), we proposed to establish testing frequency requirements for the use of untreated surface water for purposes that are subject to the requirements of proposed § 112.44. As proposed, if the untreated surface water is from any source where a significant quantity of runoff is likely to drain into the source (for example, a river or natural lake), then you must test the water at least every 7 days during the growing season (proposed § 112.45(b)(1)). If the untreated surface water is from any source where underground aquifer water is transferred to a surface water containment constructed and maintained in a manner that minimizes runoff drainage into the containment (for example, an on-farm manmade water reservoir), then you must test the water at least once each month during the growing season (proposed § 112.45(b)(2)). In proposing these testing frequencies, we tentatively divided untreated surface water into two categories based upon their potential to be adversely affected by runoff and the degree to which you reasonably could be expected to exercise protection and control over them. We tentatively concluded that runoff is the most important variable among the various environmental factors that may affect the microbial quality of surface water, because it has the potential to increase the number of pathogens in the water column if its origins include human, livestock or wildlife feces and because it has the potential to increase the amount of suspended sediments, which are likely to harbor pathogens. We also considered other factors, such as precipitation and its effects ( e.g., discharge and flow rate) along with temperature, which are common factors reported to affect the microbial quality of watersheds with agricultural land inputs. However, we did not propose a surface water testing frequency based on these factors because such an approach would require full characterization of its effects on the quality of surface water sources that are not likely to be generally useful across all farms, States, or regions (78 FR 3504 at 3571). We also noted that our approach to testing untreated surface water was to propose practical intervals of testing both because they are likely to capture transient events that may degrade quality and because they are useful regardless of geographic location. The sampling and testing frequencies we proposed in § 112.45(b) are the minimum that we tentatively concluded provide sufficient information concerning your source surface water quality for you to use in determining the method of application for which the water is safe and of adequate sanitary quality. We asked for public comments on our proposed testing frequencies, including any alternative approaches and examples where testing should be more or less frequent based on your experience or observation, and specifically if you believe that surface waters can be thoroughly characterized when tested at frequency less than that proposed in § 112.45 (78 FR 3504 at 3571). a. Relevant Comments. We received a number of comments on our proposed requirements for frequency of water testing, many of which voiced concerns and requested that FDA reduce the required testing frequencies and apply a flexible approach that considers the specific risks associated with the particular source of water and its use. Comments related to the frequency of water testing highlighted various issues, including the following: (1) Commenters recommended that FDA should employ and allow the use of risk-based testing strategies that account for the variability in risk associated with the specific source of water and its use. For example, commenters noted that the proposed testing frequencies do not consider the wide range of sources of agricultural water, which include municipal water to rural rain water catchment. Commenters also noted that frequent testing is either not necessary or does not provide meaningful information where there is inherently high variability in water quality due to rainfall or other natural events. Commenters stated that microbial growth and survival varies significantly by region and water source, and some open water sources have sufficient microbiological stability that weekly testing is unnecessary. In addition, commenters argued that the testing frequency requirements should recognize the reduced risk (and consequently, less frequent need for testing) associated with proper design and maintenance of the water source to encourage growers to implement preventive measures; (2) commenters stated that there is a need to incorporate flexibility into the testing frequency requirements so growers can determine appropriate frequencies, considering factors specific to their source of water and its use. For example, commenters asserted that testing frequencies should be tailored for farms using short-term or intermittent irrigation. In addition, some commenters stated that an assessment of risks associated with ground water should be farm-based because not all ground water is equal or merits the proposed testing frequency, and that FDA must permit alternative practices for water testing based on sound science; (3) commenters suggested that appropriate testing frequencies should be determined depending on historical test results. Commenters maintained that a more effective approach than the one proposed by FDA would be to take baseline samples to determine water quality and then schedule routine future testing based on the results of the baseline testing; (4) commenters argued that scientific data to support the proposed testing frequencies are lacking. For example, commenters opposed the specific requirements related to testing of untreated surface water in proposed § 112.45(b), and asserted that general water testing requirement in proposed § 112.45(a) to test agricultural water at the beginning of the growing and every 3 months thereafter during the growing season, coupled with the requirements in proposed § 112.42 to regularly inspect and maintain agricultural water systems, is adequate. Commenters who opposed the weekly testing requirement in proposed § 112.45(b)(1) pointed out that, although they acknowledge the need to test surface water sources more frequently than ground water sources, there is no basis for the proposed weekly testing of untreated surface water. One commenter also pointed out that a WHO analysis of tolerable risk for irrigation water determined that harvesting 5 days after last irrigation has a significant reduction in contamination. Other commenters argued that human pathogens do not survive well on produce in the field and, therefore, contamination that occurs early in a growing season may not survive to harvesting, such that a requirement to test at the beginning of each growing season would be of no value. Some commenters requested more clarity regarding the frequency of testing water that is used in harvest and post-harvest activities, and the data that FDA used to determine the adequate testing frequency for such use of water. Commenters also urged FDA to revisit the scientific data supporting the testing intervals and validate the quality of those data. Still other commenters encouraged FDA to create a separate rule or guidance on testing frequency requirements after further research is completed; and (5) commenters argued that the proposed testing frequencies would pose an undue financial burden without providing clear public health benefits. Commenters strongly opposed the weekly testing frequency, in particular, and stated that farms do not have the necessary resources or facilities to accommodate such frequent testing, and some growers would have to ship their water samples to testing laboratories. Some commenters also noted that many growers use more than one pond for irrigation and using up to four ponds is not uncommon, such that costs of testing could become prohibitively expensive. One commenter estimated that the total cost associated with water testing requirements could amount to about $11,550 annually (including costs of labor and laboratory testing). Another commenter urged FDA to explicitly permit growers to use water testing data compiled by other entities. According to the commenter, municipalities in New Hampshire routinely test E. coli levels for recreational purposes, and it would be unnecessary to require growers to test the same water source for the same pathogens separately. Commenters also recommended specific alternative testing frequencies in lieu of our proposed provisions. Some commenters mentioned that a more prudent testing requirement would be within a timeframe closer to harvest, while others suggested that it would be beneficial to require water testing at the outset for a new operation or when a new water source is first brought into use. Other notable suggestions included seasonal water sampling, or using the current USDA's Good Agricultural Practices requirements for testing surface waters at the beginning and the peak of the growing season and at harvest time. Conversely, a few commenters agreed with the testing frequencies that we proposed, stating that the proposed schedule of water testing ensures the safety of water initially and during growing, harvesting, and post-harvest activities. Overall, a majority of the concerns with the proposed testing frequencies centered on the financial burden imposed on farms, in particular, under a weekly testing requirement; that FDA did not provide scientific data in support of the proposed testing frequencies; and the need for a more flexible approach accounting for the variability in water quality associated with various water sources and the particular use of the water during growing, harvesting, or post-harvest activities. b. FDA's Consideration of Comments. As noted above, a key objective of our proposed approach to water testing was to establish a testing frequency sufficient to adequately characterize the quality of the agricultural water such that the information could lead farms to make informed and appropriate decisions about its use and/or the need for any appropriate corrective actions, prior to such use. Commenters generally agreed with our intent to characterize the quality of the water source, but argued that the frequency intervals proposed were too short; and, as a consequence, would require more tests (and associated costs) than necessary to accomplish the desired outcome, without a commensurate gain in public health benefit. In view of comments received, we reviewed our previous proposed frequencies to characterize the quality of untreated surface water and untreated ground water sources. Taking into account comments received, currently available information, and upon further analysis, we are proposing certain amendments to proposed §§ 112.45 that, collectively, result in a proposed tiered approach to testing untreated surface water and untreated groundwater. The proposed approach would allow farms to make decisions about safe use of available water sources prior to the beginning of the next growing season; adjust testing frequencies dependent on long-term test results; and ultimately reduce the required frequency of testing. In the case of both untreated surface water and untreated ground water, we are proposing to more narrowly focus the period of characterization of water quality to those when the risk is greatest, i.e., during periods when agricultural water is used immediately prior to harvest. Currently available information indicates that the risk to consumers is greater in relative terms when produce contamination via agricultural water occurs closest to consumption. That is, agricultural water used early in the growing season ( e.g., seeding, plant establishment) generally has less inherent risk associated with its use than water used in harvest ( e.g., field wash) or post-harvest activities ( e.g., washing, cooling). Requiring that water characterization focus on periods when the risk is greatest reconciles public comments with the scientific literature on the relative risks associated with the timing of use of agricultural water. This approach is supported by the discussion above concerning die-off rates between application of water and harvest. With die-off rates of 0.5 log or greater per day the impact of water quality more than a couple of weeks prior to harvest is minimal. We expect this time period ( i.e., immediately prior to harvest) to be variable and dependent on the crop and length of time harvest activities are performed. It is reasonable to conclude that it would include periods immediately prior to active harvest of one commodity or variety, even though another continues to mature but is not yet ready for harvest. To permit farms to tailor their sampling of water to the unique circumstances relevant to their crop(s) and practices and conditions on their farm, we are proposing as a requirement that the samples required to be collected include those “collected during a time period(s) as close as practical to harvest.” We recognize that the timing of the use of agricultural water using a direct application method varies by crop, region, season, and/or from year to year. By using the term “practical,” we intend to convey that agricultural water should be collected for analysis when, during the characterization or verification period, agricultural water is applied to covered produce, and not that samples would be collected from the source water when it is not being applied to the crop. Timing of the samples should be such that the last applications of agricultural water prior to harvest are targeted, again recognizing that in some circumstances such applications may not be preplanned ( e.g., early frost or unusually hot, dry weather). Further, timing of sample collection should occur in the time period during growing and near harvest, and be designed to represent events that can reasonably be expected to both impact water quality ( e.g., rainfall, high river stage, wildlife and domesticated animal movement through upstream water systems) and occur in the time period during growing and/or near harvest. At this time, we are not proposing to further specify an appropriate time period prior to harvest for sampling. We seek comment on whether it would be practical to require sample collection during a certain time period(s) such that the test results based on such samples would be available in sufficient time to determine any changes to water quality and, if necessary, adjust harvesting times accordingly or take other corrective actions. i. Tiered approach to testing untreated surface water. We are proposing to amend proposed § 112.45(b) to establish a new proposed tiered approach to testing untreated surface water that is used for the growing of produce (other than sprouts) using a direct application method. As amended, proposed § 112.45(b) would establish that if you use untreated surface water for purposes that are subject to the requirements of proposed § 112.44(c), you must take the following steps for each source of the untreated surface water: (1) Conduct a baseline survey to develop a water quality profile of the agricultural water source. (i) You must conduct a baseline survey in order to initially develop the water quality profile of your water source. You must determine the appropriate way(s) in which the water may be used based on your water quality profile in accordance with § 112.44(c)(1) through 112.44(c)(3). (ii) The baseline survey must be conducted over a minimum period of 2 years by calculating the GM and the STV of generic E. coli (CFU per 100 mL) using a minimum total of 20 samples, consisting of samples of agricultural water as it is used during growing activities using a direct water application method, collected during a time period(s) as close as practical to harvest. The water quality profile initially consists of the GM and STV of generic E. coli calculated using this data set. (iii) You must develop a new water quality profile: (A) At least once every 10 years by recalculating the GM and STV values using a minimum total of 20 samples collected during your most recent annual surveys (which are required under paragraph (b)(2) of this section); and (B) when required under paragraphs (b)(2) and (b)(3) of this section. (2) Conduct an annual survey to verify the water quality profile of your agricultural water source. (i) After the baseline survey described in paragraphs (b)(1)(i) and (b)(1)(ii) of this section, you must test the water annually to verify your existing water quality profile to confirm that the way(s) in which the water is used continues to be appropriate. You must analyze a minimum number of five samples per year, consisting of samples of agricultural water as it is used during growing activities using a direct water application method, collected during a time period(s) as close as practical to harvest. (ii) If the GM and/or STV values of the [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2014-22447. Public record. Not legal advice.