UCMR3 Results and Implications for UCMR4 NC AWWA-WEA August 2016 Thank you all for coming today and for the opportunity to present this information regarding UCMR. My name is Joe Mattheis. I am based in Raleigh and work with Eurofins Eaton Analytical, Inc. supporting our South Bend, IN and Monrovia, CA Labs as both an Accounts Manager (with responsibility for sales) and Analytical Services Manager (project manager responsible for ordering, scheduling, reporting and invoicing) for NC based customers. I’ll present information regarding UCMR3 results, and a look forward to UCMR4. Monrovia, California South Bend, Indiana

Purpose of UCMR “To collect occurrence data for contaminants suspected to be present in drinking water but don’t have health-based standards set under the Safe Drinking Water Act (SDWA).”* *EPA Fact Sheet EPA 815-F-12-003 In 1996 Congress amended the Safe Drinking Water Act requiring EPA to develop a new list of up to 30 unregulated contaminants every 5 years for evaluation and subsequent regulatory consideration. EPA’s stated UCMR Purpose : “To collect occurrence data for contaminants suspected to be present in drinking water but don’t have health-based standards set under the Safe Drinking Water Act (SDWA).”* UCMR is part of the Safe Drinking Water Act

UCMR3 – A LOT of Data EPA has released data ~ Quarterly, the last set in April 2016 Data now represents ~61,000 samples (~36,000 entry point samples and 25,000 Maximum Residence time samples) from multiple labs. It is about 1,024,000+ points. Data from our labs accounts for 40% of those results. Represents about 4863 PWS for List 1 and about 1186 PWS for List 2. The original 3 year sampling schedule ended 12/31/2015. EPA has released data ~ Quarterly, the last set in April 2016 Data now represents ~61,000 samples (~36,000 entry point samples and 25,000 Maximum Residence time samples) analyzed at multiple labs. It is about 1,024,000+ points. Data from our labs accounts for 40% of those results. Represents about 4863 PWS for List 1 and about 1186 PWS for List 2. The original 3 year sampling schedule ended 12/31/2015.

Data Represents a Good Mix of GW (2200) and SW (7200) Sites Public Water supplies utilizing both Ground Water and Surface Water sources participated. Groundwater sites are marked in Green Surface water sites are marked in Pink

Occurrence Frequency Has Not Changed as More Data Published Overall patterns of occurrence have not changed that much since the first NCOD data release. % of PWS with Detects Contaminant 10-13 1-14 4-14 7-14 10-14 6-15 10-15 4-16 1,4-dioxane 19% 20% 22% PFOS 1.5% 1.6% 1.8% 2.0% 1.9% Vanadium 75% 77% 74% 70% 72% 73% Hex Chrome 89% 90% 87% 88% Testosterone 4.0% 5.0% 4.6% 4.7% 4.5% 4.8% Over the course of the 3 year sampling schedule, the percentage of Public Water Supplies with results occurring above the Minimum Reporting Level has remained consistant for most contaminants measured.

% of total results >Reference Concentration Out of 28 Chemical Contaminants, There are Really Only A Few Significant Ones Contaminant % of Results > MRL % of total results >Reference Concentration % of PWS with results > MRL % of PWSs with results >Reference Concentration 1,2,3-trichloropropane 0.69% 0.7% / 0.5%1 1.32% 1.3% / 1.1%1 1,1-dichloroethane 2.29% 0.003% / 0%1 4.93% 0.02% / 0%1 HCFC-22 2.26% - - 5.75% Halon 1011 1.76% 0% 6.23% 1,4-dioxane 11.56% 3% / 0%1 21.90% 7% / 0%1 Vanadium 60.14% 2.7% 73.61% 3.3% Molybdenum 40.58% 0.2% 51.62% 0.8% Strontium 99.76% 2.8% 100.00% 5.7% Chromium 50.74% 0.002% 74.08% 0.02% Chromium-6 75.60% 89.33% Chlorate 55.03% 15.6% 68.92% 38.1% Testosterone 0.57% 4.89% 4-androstene-3,17-dione 0.84% - 6.16% Here we’re comparing percentages of total results: greater than or equal to the MRL in the first column greater than or equal to Reference Concintrations (10 to the -6 cancer risk, and 10 to the -4 cancer risk) in the second Column And Percentages of PWS with results: greater than or equal to the MRL in the third column greater than or equal to Reference Concintrations (10 to the -6 cancer risk, and 10 to the -4 cancer risk) in the fourth Column Yellow – frequently occurring, but natural or not at significant levels. Red – frequently occurring and/or significant # of samples/PWS >HRL Yellow – frequently occurring, but natural or not at significant levels. Red – frequently occurring and/or significant # of samples/PWS >HRL.

Chlorate is Present at Significant Levels in over 15% of Samples Nationwide 37% of PWS EXCEED the HRL (210 ug/L). PWS are using hypochlorite more than gaseous chlorine post 9/11. Bulk hypochlorite is a significant source of chlorate; but so is onsite generation. Chlorate can be easily controlled in bulk hypo. >5% of plants using onsite generation of hypo for chloramines have chlorate > WHO limit

Metals are More of a Groundwater Issue Than a Surface Water Issue Although metals are detected almost as frequently in systems with surface water sources as in groundwater systems… Concentrations tend to be significantly higher in systems with groundwater sources.

Hexavalent Chromium is Widespread, But High Values Are Isolated

Strontium Is “Regionally” High Compared to the New HRL (1500 ug/L) The lowest measured concentration is 10X the MRL.

1,4-Dioxane is Widespread, But High Values Are Clustered Detected in 12% of samples nationwide ~3% exceed the 0.35 ug/L HRL

1,4-Dioxane is Both a GW and SW Issue Many of the surface water hits are in the Southeast and likely represent point sources

Sidebar – 1,4-Dioxane Uses Stabilizer for chlorinated solvents (1,1,1-trichloroethane) Used in the production of many products: Paint Strippers, Varnishes and Waxes Dyes Greases Wetting agent in textile industry Purifying agent in pharmaceutical production

Sidebar – 1,4-Dioxane is Present as a Contaminant in Many Things Vine-ripened tomatoes (contaminant in applied pesticides) By-product of polyethylene terephthalate (PET) plastic manufacturing Impurity in antifreeze and aircraft deicing fluids (HCAA) Campaign for Safe Cosmetics found 1,4-Dioxane in 22% of cosmetic products!! (Not required to be listed as an ingredient because it is a contaminant.) UBIQUITOUS!!

Sidebar – 1,4-Dioxane Concerns EPA listed as a probable carcinogen (B2) Highly miscible and mobile in groundwater (think MTBE) Not biodegradable Does not readily evaporate from surface waters Seems likely to be regulated but at what level? (0.3 to 5 ug/L action levels already in place in some states)

Volatile Organic Compound Occurrence Overall, about 5% of samples have 1 or more VOC detections (minimal co-occurrence) As expected, almost all the hits are GW samples. Most common detections: 1,1-DCA (3%) Chlorodifluoromethane aka HCFC-22 (2.2%) Bromochloromethane aka Halon 11 (2.0%)

VOC Occurrence is Very Regional and Very Low Overall

Perfluorinated Compounds(PFCs) Detections are Infrequent (N ~36,000) Frequency of Detection as % of samples % PWS w Hits 99th % conc Max conc (ug/L) HRL (ug/L) PFOS 0.8% 2.0% ND 1.8 0.07 PFHxS 0.6% 1.2% 0.68 -- PFHpA 0.7% 1.6% PFOA 1.0% 2.1% 0.35 Detections in ~ 20 states; not necessarily consistent hits over time. GW system frequency and levels are higher than SW Many of the hits are non-CCL3 PFCs (only PFOA and PFOS are on the CCL3 list).

UCMR 3 PFCs by Method 537 UCMR 3 PFCs Data Summary - April 2016 Contaminant MRL (μg/L) Reference Concentration (μg/L) Total number of results Number of results > MRL Number of results >Reference Concentration % of total results >Reference Concentration Total number of PWSs with results Number of PWSs with results > MRL Number of PWSs with results >Reference Concentration % of PWSs with results >Reference Concentration PFOS 0.04 0.07 36149 285 119 0.3% 4864 94 46 0.9% PFOA 0.02 36148 354 31 0.09% 108 13 PFNA NA 36150 19 - - 14 PFHxS 0.03 204 55 PFHpA 0.01 231 84 PFBS 0.09 18 - 8 EPA - April 2016 UCMR 3 Data Summary for Chemical Contaminants  

Sidebar- EPA DW Health Advisory - PFCs Help local water systems, state, tribal and local officials take steps to address PFOA and PFOS Establishes individual or combined concentrations in DW of PFOA and PFOS at 70 parts per trillion EPA Method 537 is recommended for Analyses EPA issued Health Advisories in May of 2016 for PFOA and PFOS. Establishes individual or combined concentrations in DW of PFOA and PFOS at 70 parts per trillion EPA Method 537 is recommended for Analyses

Perfluorinated Compound Hits Likely Represent Local Industrial Sources

PFC6 by Method 537 Parameter CAS # Reporting Limit UCMR3 MRL Federal Limit  Perfluorobutanesulfonic acid (PFBS)  375-73-5  9.0 ng/L  90.0 ng/L  N/A  Perfluoroheptanoic acid (PFHpA)  375-85-9  1.0 ng/L  10.0 ng/L  Perfluorohexanesulfonic acid (PFHxS)  355-46-4  3.0 ng/L  30.0 ng/L  Perfluorononanoic acid (PFNA)  375-95-1  2.0 ng/L  20.0 ng/L  Perfluorooctane sulfonate (PFOS)  1763-23-1  4.0 ng/L  40.0 ng/L  70 ng/L* (LHA)  Perfluorooctanoic acid (PFOA)  335-67-1 Slide shows South Bend Current RLs. Working toward 2.0 ng/L for all compounds at both test sites.

PFCs - Field Blanks Guidance Method 537 requires the use of Field Blanks where reagent water sent from the lab is transferred into a blank bottle at the sampling location. Field blanks require analysis when the associated field sample has any detection. For post-UCMR3 confirmation analysis, most of these sample sources are anticipated to have detects which will require us to analyze Field Blanks for almost every sample.

Hormones Present a Conundrum- Non CCL Dominate . ~9,000 samples (but only ~960 PWS, 3600 sites) 133 hits—maximum values from 1 to 5 part per trillion 4-androstene-3,17-dione (74 hits 68 < 1 ppt) max 1.9 ppt Testosterone (52 hits – 50 < 1 ppt) max 5.3 ppt 17-alpha-ethynylestradiol (3 hits) max 1.6 ppt estriol (1 hit) max 1.1 ppt 17b-estradiol (3 hits) max < 1 ppt Neither of the most frequently detected analytes are on the CCL 3 List. (95% of hits)

Hormones Are Infrequent (25 States), But Some In Unexpected Areas Almost all of the hormone data seem to be one time hits (e.g. very problematic to explain)

Implications of UCMR3 for UCMR4 How low should we go? (are we chasing 0) What frequency of detection is meaningful? What ancillary data (metadata) are useful? When does the distribution system matter? How do we balance CCL compounds vs analytical method target lists? How frequently do we sample? When do we sample? There are lots of questions regarding what is necessary, and we will briefly raise questions about these, but the main focus is on whether the number of samples being collected and analyzed is overkill

Detection Does Not Mean a Health Risk 1990 UCM 1 UCM 2 SDWA (1996) 2000 UCMR 1 UCMR 2 2010 UCMR 3 27 years UCMR4 (2018) From Via, 2015

Reporting Limits Should Be Reviewed Critically and Not Just be Formulaic Have we engaged in overkill by relying on the Lowest Conc Min Reporting Level (analytical methods capability) to set reporting limits? Example: Sr (lowest detected value > 10X MRL) Why look so low if we are just going to compare to reference levels (HRLs) like we did in past UCMRs?

What’s the threshold occurrence for potential regulation? While Regulatory Determinations are NOT the Same as UCMR, They ARE Inter-related. What’s the threshold occurrence for potential regulation? % of detections? % exceeding current HRL? Examples: 1,4-dioxane is detected in 20% of PWS with 7% of PWS exceeding 10-6 HRL Strontium exceeds HRL in 5% of PWS

Are We Really Looking for the Right Compounds? Are we chasing zero? Other than 1,4-dioxane, most of the organics are still mostly ND, even at ultra low MRLs Is the existing CCL really the best source for determining what to monitor? “Frequently” detected hormones not on CCL And you won’t find what you don’t look for…

The Current UCMR Framework is Overly Rigid We are still chasing ultra-low numbers because we can. We have completed LC MRL determinations for potential UCMR4 methods for EPA How do you fit cyanotoxins into the UCMR framework? Sporadic in occurrence Triggered monitoring makes the most sense UCMR approach would estimate low Triggered approach could estimate high

Metadata Can Help Utilities With Interpretation Disinfection processes CAN impact some contaminants (e.g. chlorate), but if the data are not detailed, it doesn’t help utilities to understand and control formation. Requires further detailed studies Source water data can be critical when analytes may be formed/removed in the treatment process. e.g. Cyanotoxin strategies depend on source info.

What About by Population? There are definite differences in frequency of detection as a function of system size. Hence any changes in UCMR4 should still include multiple population categories.

Conclusions- UCMR3 Results Implications for UCMR4 CALIFORNIA CECs Conclusions- UCMR3 Results Implications for UCMR4 As long as we look for ultra-low concentrations, we are going to get hits and create communications challenges for utilities. UCMR4 could be much more valuable if we consider including source waters in the monitoring and think carefully about when DSMRT monitoring makes sense. There needs to be a better QC feedback loop to ensure meaningful data.

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Cyanotoxins - Methods ADDA Elisa (assay kit) – Total Microcystins Method 544 (SPE LC/MS/MS) – Microcystins LA,LF,LR LY, RR,YR, and Nodularin Method 545 (LC/ECI-MS/MS) - Anatoxin A and Cylindrospermopsin

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Fourth Unregulated Contaminant Monitoring Rule _ Monitoring the Occurrence of Unregulated Drinking Water Contaminants _ US EPA.html

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Metals Method 200.8 (ICP/MS) Germanium Manganese

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Organics Method 525.3 (SPE GC/MS) Tebuconazole Alpha-hexachlorocyclohexane Cis & Trans Permethrin Chloropyrifos Tribufos Dimethipin Ethoprop, Oxyfluorfen, Profenofos

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Organics Method 552.3 (GC/ECD) HAA5 HAA9 HAA6Br

UCMR4 – Proposed Analytes CALIFORNIA CECs UCMR4 – Proposed Analytes Organics - Continued Method 541 (GC/MS) 1-Butanol 2-Propenol Method 530 (GC/MS) Butylated hydroxyanisole Quinolone

Eurofins Eaton Analytical, Inc. Any Questions? Joe Mattheis JosephMattheis@eurofinsus.com Raleigh, NC 919-376-7978 Andy Eaton, PhD, BCES andyeaton@eurofinsus.com 626.386.1125 Eurofins Eaton Analytical, Inc. www.eurofinsus.com/eaton