2020.10.05_CCOA.p1_DEQResponseToMassLoadingProtocols

From:Grzyb, Julie
To:Compton, Christel E
Cc:Garon, Kevin P; Smith, Danny; Holman, Sheila
Subject:Re: Department comments on Mass Loading Protocols
Date:Monday, October 5, 2020 11:26:34 AM
Attachments:AC75751.pdfAC75752.pdfAC75753.pdfAC75754.pdfLock3 Map_8_4_2020.jpg
Attachments and picture referred to in comment 8.
Julie
Julie A. Grzyb, Deputy Director
Division of Water Resources 1617 Mail Service Center, Raleigh, NC 27699-1617919/707-9147 (wk); 336-210-8454 (cell))
From: Grzyb, Julie <[e-mail]>
Sent: Monday, October 5, 2020 9:21 AM
To: Compton, Christel E <[e-mail]>
Cc: Garon, Kevin P <[e-mail]>; Smith, Danny <[e-mail]>;
Holman, Sheila <[e-mail]>
Subject: Department comments on Mass Loading Protocols
October 5, 2020 Christel,
DEQ did a review of the Mass Loading protocols submitted by Chemours and we look forward tohaving the opportunity to discuss the basic equation concept and specifically what factors will be
included in the initial Baseline Calculation and the timeframe these calculations are based upon. Inaddition, we have the following questions/concerns with the protocol that we are hoping can beaddressed/clarified during the meeting.
1). For the Cite Conveyance Network the facility suggests the use of 3.5 days composite samples and24-hour composite sampling. The 3.5 days sampling is inconsistent with our regulations and
underestimates the load, only 24-hour composite data should be used in the calculations.
2) The document mentions the phrase “estimated PFAS concentration” several times. Please clarifythe use of that term and note that measured PFAS concentrations are expected to be used in the
calculations. For example, = is the measured or estimated total PFAS concentration.

3) The protocol states how PFAS concentrations from the site-conveyance network may be estimatedif there are data gaps, and for the intake, estimated concentrations will be determined from
Transport pathways 1 (upstream at mile 76), 2 (Willis Creek estimated modeled load), and 3 (aerialdeposition). Actual concentrations will also be measured monthly with a 24hr composite sample at
the intake. It would be better to take weekly 24hr composite samples at the intake to determineloading from the site-conveyance network. This would also provide more data comparison to refinethe loading as estimated from Pathways 1-3 since two of those inputs are modeled loading and notactual measurements. The last sentence of 2.4 says; For the purposes of these calculations, flows at
the river intake are assumed to be equal to dry weather Outfall 002 flows. This seems to indicatethere is no consumptive loss from the variety of processes used at the industrial site. Shouldn't thewater balance include water lost from hauling process wastewater and evaporative loss from variousprocesses?
4) The document suggests the use of a 60-day rolling average. This is inconsistent with our standard
procedures, a 30-day rolling average should be used.
5) The list of PFAS parameters that will be analyzed should match the list contained in the recentlyissued NPDES permit – NC0089915 and as compromised during CO negotiations.

6) The CO states that for Outfall 002, Chemours shall also conduct 24-hour composite samplingwithin 24 hours of any rain event predicted two days before with at least a 70% likelihood to be of 1 inch or greater over a 24-hour period. The weather service used for theprecipitation prediction should be stated and approved by the department.

7) OLDOF-1, this sampling location is approximately 500' upstream from the confluence with theCFR. Sampling should take place closer to the mouth unless it can be demonstrated that the CFR isinterfering with sampling between the designated location and the 'mouth' of the old outfall 002channel.

8) The RRO discovered two seeps at the boat ramp below L&D#3 draining into the river. Theysampled the seeps and the results are attached. The loading from these seeps should somehow becaptured in the sampling plan. 9) Section 4.2: Equation 6 is used to estimate the mass of PFAS at the intakes for the two watertreatment plants downstream of Chemours (Bladen Bluffs and Kings Bluff). We understand the
equation, except Chemours does not explain why the flow recorded at the W.O. Huske Dam needs tobe adjusted for travel time for the Bladen Bluffs plant using Equation 7.

10) Chemours should provide an example dataset for the equations presented in the protocol, alongwith the spreadsheets planned for calculating the mass loads and mass discharges (the spreadsheetscould be populated with the example dataset). This would further aid in understanding the formulasand also provide a mechanism to check for errors.

11) Section 3.1. The variables “j” and “k” are not defined in Section 3.1.

12) Section 3.2 provides Equation #2 and has a variable “F” that needs further explanation. Is thefactor a constant or will it change each time?

13) Section 4.4 does not include loading from the floodplain deposits adjacent to the Cape FearRiver: The PFAS mass loading model does not account for PFAS loading of the river fromgroundwater contamination or contaminants in soil of the vadose zone. Loading of PFAS from thefloodplain deposits to the river will occur until the extraction well system and/or barrier wall areconstructed. Chemours should clarify why PFAS loading from the floodplain deposits are notincluded.

14) Section 4.4.1: The equation labelled “[1]” should perhaps be labelled the 9th equation presentedin the body of the text. Chemours also does not have an equation #8 in the protocol.

15) Section 4.4.4: Pathway TP-5 is the upwelling of PFAS from the Black Creek Aquifer into theRiver at 8 segments of the river bank. Mass discharge for each segment will be determined using thecross-sectional area, the hydraulic gradient for the groundwater surface near the river and thehydraulic conductivity (K) estimated from slug tests performed on individual wells screened in theBlack Creek Aquifer. In general, K values estimated from multiple well aquifer pumping tests arepreferred over those estimated from slug tests. Kruseman and de Ridder (1994) note that “…slugtests cannot be regarded as a substitute for conventional pumping tests” and that a slug test “…determines the characteristics of a small volume of aquifer material surrounding the well, and thisvolume may have been disturbed during the drilling and construction.” Chemours has indicated thatpumping tests are planned during the design phase for building of the groundwater extraction systemand barrier wall.
16) Section 4.4.6 references Appendix D, which has an equation for the total PFAS discharged fromgroundwater adjacent to Chemours (east side of river) and downstream of Chemours within a 1-miledistance of the river. Wells with contaminated groundwater were detected within the 1-mile width oneither side of the river. The equation includes a unitless “scaling factor” which is a ratio of the lengthof the river adjacent to the site + downstream divided by the length of the river upstream.

17) Based upon our review, it appears more appropriate to input “land areas” into the ratio instead of“river length” units. Chemours includes a figure that shows the off-site land areas underlain byimpacted groundwater and the river lengths. The figure shows a width that extends greater than 1mile on either side of the river, based on comparing the mapped widths to the bar scale.

18) The Draft Mass Loading Assessment Summary dated Dec. 6, 2019 determined hypothetical massloading scenarios of HFPO-DA as reductions occurred at the facility. To continue this modeling andconfirm these predictions PFAS samples should be taken below Outfall 002 (outside the mixing zoneas determined by the dilution model) and below Outfall 003 when samples are taken upstream.

We look forward to discussing these concerns/questions on Wed. Respectfully, Julie

Julie A. Grzyb, Deputy Director Division of Water Resources 1617 Mail Service Center, Raleigh, NC 27699-1617 919/707-9147 (work)