Policy Regarding Alteration Of Groundwater Samples Collected For Metals Analysis (TAGM - 4015)

Issuing Authority: Michael J. O'Toole, Jr.
Title: Director, Division of Environmental Remediation
Date Issued: September 30, 1988

A. Objective

The purpose of this Technical and Administrative Guidance Memorandum (TAGM) is to clarify the New York State Department of Environmental Conservation's (NYSDEC) Division of Hazardous Waste Remediation's policy of non-alteration of groundwater samples collected for metals analysis, as well as provide guidance for situations where the policy may be amended. This TAGM has been developed for the most commonly encountered form of a alteration (as defined in Section B), which is filtration (as defined in Section B), but the policy applies to all forms of groundwater sample alteration which may be encountered.

(*Note: This guidance is for situations involving groundwater samples collected for metals analysis only. It does not apply to surface water, waste or samples at active sites or State Pollution Discharge Elimination System discharges where the proper handling of samples may be dictated by other guidance or protocols.

Department policy regarding water samples collected for the analysis of organic compounds is expressly stated, in an Executive Memorandum dated May 3, 1985, from Commissioner Williams to Executive Staff, Division Directors, and Regional Directors, as follows: "Water samples utilized in the assessment, investigation, remedy, study, construction, monitoring or any other activity shall not be altered prior to analysis.")

B. Definitions

"Alteration" - changing the sample in any way other than adding a preservative, such as nitric acid, to lower pH. Examples of alternation include, but are not limited to: filtering, settling and decanting, centrifuging and decanting, and acid extracting.

"Filtration" - the filtering of a groundwater sample collected for metals analysis, through any membrane, fabric, paper or other filter medium irrespective of pore size, in order to remove particulates from suspension; it is to be done in the field, at time of collection, prior to preservation.

"Preservation" - the preserving or "fixing" of metal ion constituents in a groundwater sample, so as to avoid precipitation, biodegradation, or loss of the ions from the sample. Methods utilized are as listed in the New York State Contract Laboratory Protocol (NYSCLP). Generally they involve the reduction of sample pH to <2 with HNO3 (nitric acid).

"Well Development" - the application of energy to a newly installed groundwater monitoring well in order to allow the natural hydraulic properties of the screened formation to return, thus allowing water to flow more freely to the well. It may also allow the removal of any formation material that may have infiltrated the sandpack and/or well during installation.

"NTU" - nephelometric turbidity unit; this is the unit by which turbidity is measured and discussed.

"50 NTU maximum" - target turbidity level for development and sampling of groundwater monitoring wells; any higher and the sample may be considered unacceptable. This is the level above which laboratory analysis problems may occur.

"Split Sample" - a single sample divided into aliquots.

"ARAR's" - Applicable or Relevant and Appropriate Requirements, as defined by the Superfund Amendments and Re-authorization Act of 1986 (Section 121).

"RCRA" - Resource Conservation and Recovery Act.

C. Existing Guidance

In accordance with 40 CFR 136, Guidelines Establishing Test Procedures for the Analysis of Pollutants Under the Clean Water Act; Final Rule and Interim Final Rule and Proposed Rule, Section 40 CFR Part 136.3, Table IB, Note 3, states a sample collected "for the determination of total metals... is not filtered before processing," whereas Section 40 CFR Part 136.3, Table II, Note 7, states for samples collected for dissolved metals analysis "should be filtered immediately on-site before adding preservative...".

Information provided in the RCRA Groundwater Monitoring Technical Enforcement Guidance Document, 1986, USEPA-OWPE, OSWER-9950.1, P. 114, is as follows:

"Metallic ions that migrate through the unsaturated (vadose) and saturated zones and arrive at a groundwater monitoring well may be in the well. Particles (e.g., silt, clay), which may be present in the well even after well evacuation procedures, may absorb or adsorb various ionic species to effectively lower the dissolved metal content in the well water. Groundwater samples on which metals analysis will be conducted should be split into two portions. One portion should be filtered through a 0.45-micron membrane filter, transferred to a bottle, preserved with nitric acid to a pH less than two, and analyzed for dissolved metals. The remaining portion should be transferred to a bottle, preserved with nitric acid, and analyzed for total metals. Any difference in concentrations between the total and dissolved fraction may be attributed to the original metallic ion content of the particles and any sorption of ions to the particles."

Information provided in the RCRA Comprehensive Groundwater Monitoring Evaluation Document, March 1988, USEPA-OWPE, Directive 9950.2, p.22, is as follows:

"Metals

1. Samples collected for metals analysis should be split into two samples. One portion filtered through a 0.45 micron filter for dissolved metals and the second portion remaining unfiltered for total metals analysis. Samples should be filtered as soon as possible to minimize the impacts of pH and Eh changes."

Present Division guidance on filtering of groundwater samples collected for metals analysis consists of memorandum from N. Nosenchuck, Director, Division of Solid and Hazardous Waste, to P. Buechi, Region 9 Solid Waste Engineer, dated January 21, 1987, in response to a memorandum from P. Buechi to N. Nosenchuck dated December 4, 1986, in which Mr. Buechi inquired about the policy of the Division regarding the filtration of samples collected for metals analysis. The response memorandum stated the Division policy as follows:

"The Division of Solid and Hazardous Waste guidance is that samples submitted for metals analysis not be filtered prior to analysis. This general guidance, however, can be amended on a case-by-case basis, when it can be shown that a compelling reason exists for sample filtration. Such reasons could include the modeling of water treatment processes or the determination and/or elucidation of adsorptive distribution phenomena. Ground and surface water samples collected for the purpose of generating monitoring data or for site investigation must be analyzed whole."

D. Discussion

1) Goals of Site Investigations (Remedial Investigation/Phase II)

Generally, for the State Superfund program, the reasons for taking a groundwater sample at an inactive hazardous waste disposal site are:

to determine whether or not a significant threat to health or the environment exists;
to characterize the site from the point of view of past history, present situation and any future actions that may be necessary;
to determine the nature and extent of contamination; and
to obtain sufficient data to support the choice of a remedy in the feasibility study.

2) Obtaining These Goals

First, to accomplish the above goals, the results of the sample analysis are compared with standards, guidelines and other ARAR's, as one element of determining whether or not a significant threat to human health or the environment exists. Because Part 703 groundwater standards and the State Health Department's drinking water standards are all based on samples that are not altered, samples from sites must be analyzed in their unaltered state so that a valid comparison can be made.

Second, characterization of the site, determination of the extent of contamination, and preparation of feasibility studies involve broadening the scope of work to include past and future conditions. Groundwater may be used in the future for drinking or irrigation purposes, or it may travel through the ground and intercept a well being used for potable water. Consequently, samples must be analyzed with these possibilities in mind. In order to investigate the significance of future use of groundwater at a site, groundwater monitoring wells must mimic such future use. Since groundwater is almost never filtered before being consumed or otherwise utilized, samples for site characterization must not be filtered as a standard procedure. In other words, analytical results obtained from filtered samples may not be representative of future use of the groundwater resource, and thus would not be particularly useful for evaluating the potential future toxicological impact of site contamination.

A counter-argument that has been raised is that the particulate matter in groundwater travels little distance, if at all, and therefore should not be evaluated as part of the groundwater resource that could be pumped out of the ground. The following is a rebuttal to such a counter-argument:

desorption caused by a change in water chemistry, and sometimes caused merely by changes in the concentrations of constituents, can allow metals on particles to move through groundwater even though particles themselves are remaining stationary; and
there is recent circumstantial evidence that fine (colloid-size) particles are moving through the pores of coarse-grained (sand and gravel) deposits allowing the movement of sorbed contaminants (EPA document number CERI-87-45, "Seminar on Transport and Fate of Contaminants in the Subsurface").

In addition, wells which produce water via secondary permeability (i.e. fractures in bedrock) might produce water which may have substantial suspended sediment within it. This material is not to be filtered out.

E. General Guidance for Alteration of Samples

Non-alteration of groundwater samples collected for metals analysis shall be the standard operating procedure for the Division. It is never to be assumed or written into Phase II or Remedial Investigation workplans that alteration of samples is or will be allowed or expected.
In cases where turbidity of water samples collected for metals analysis exceeds the maximum allowable level for analysis, as allowed by the Division (50 NTU), NYSDEC will review the circumstances to determine the course of action. Samples have a turbidity greater than 50 NTU's are not to be filtered as a standard procedure. Samples below 50 NTU's are never to be filtered. The rationale for any proposed filtration must be reviewed and approved by a NYSDEC QA/QC officer prior to it being effected.
When unacceptably turbid groundwater samples are produced from a well, filtration is not to be considered unless Division technical staff are certain the well was properly designed, installed, constructed, developed, maintained, and samples. Filtration is never to be the "cure" for improperly built/developed wells which produce turbid samples. Attempts should be made to repurge and/or redevelop, or replace, the well as necessary (see: Decision Flow Chart, Figure 1), and to confirm, with documentation, proper well construction and installation. The Division's technical staff who are involved must utilize best professional judgement in such cases to determine if the monitoring well has proper integrity.
(See RCRA TEGD, pp. 93-4, for discussion.) If, after a best effort at continuous development, (as agreed to by NYSDEC), it is apparent the 50 NTU's is unattainable in particular well, a decision shall be made on a case-by-case basis to:
1. install a new well at the same location (this would be based on the conclusion that the well assembly, filter pack, seals, and/or grout could be reinstalled so as to significantly reduce turbidity);
2. install a new well in a different (either vertically or horizontally) location; or
3. accept the present well as the best well attainable by reasonable means.

F. Requirements For Allowing Filtration

FIGURE 1 - Decision Flow Chart for Filtration of Groundwater Samples Collected for Metals Analysis

Filtration of properly produced groundwater samples on which analysis for metals is to be performed will be allowed only if samples of unacceptably high turbidity are unavoidable. In this case, the following protocol must be followed:

Filtration as discussed in this context involves filtering as defined in Section B of this TAGM. It should be performed consistent with the methods in the November, 1986, Environmental Protection Agency document entitled "Test Methods for Evaluating Solid Waste " (EPA-SW846).
At no time are filtered samples to be collected without an accompanying unfiltered sample. Groundwater samples are to be collected using a minimally disturbing method (i.e. low rate bladder or peristaltic pumping, bailing, etc.). Two samples will be collected, the first of which will be preserved immediately in an unaltered state, the second being filtered and preserved immediately. The turbidity of the samples should be recorded at the time of collection. If split samples are required, then both the filtered and unfiltered samples should be split.
Due to the relatively long holding time allowed for most metals, the following is recommended:
1. Analyze the unfiltered sample first.
2. If the unfiltered sample exceeds ARAR's, analyze the filtered sample.
3. If the unfiltered sample meets ARAR's, there is no need to analyze the filtered sample.
Filtration methodology must be such that changes in water chemistry of the sample are minimized. Any precipitates which may form upon removal of the sample from the well (e.g. iron floc) must not be filtered out, but redissolved by acidification/preservation. The methodology to be used must be reviewed and approved by a NYSDEC QA/QC officer prior to implementation.
Thorough documentation of the procedure(s) used is required, so that analytical results may be properly interpreted.
It may be necessary to design the analytical program to be able to answer the question whether metal contaminants are naturally occurring, or whether they were introduced through man-made activities, by analyzing upgradient and background wells by this same method. A best possible effort should be made for obtaining a "clean" or uncontaminated sample of the horizon which is being screened, so as to allow a comparison of contaminant data to naturally occurring metal ion concentrations in the aquifer matrix. This may be of critical importance, considering the potentially wide concentration ranges of various elements in native soils (Dragun, 1988).

G. Related Issues

Goals Determination
The goals for, and the type of information desired from, the groundwater sampling program must be determined prior to the commencement of any field work. If well integrity is found to exist, and turbidity is less than 50 NTU, then filtration is not considered. When considering filtration, adsorptive phenomena must also be taken into account for a proper perspective when making this decision. It must not be assumed that the groundwater and aquifer material are in equilibrium. The reactions between solid, liquid, and solution phases are complex, being defined by reduction-oxidation reactions and pH as they affect solubility and exchange reactions (Trela, 1985, Lindsay, 1979). These relationships may be altered by removal of the sample from the well, and subsequent exposure to the atmosphere (Stolzenburg and Nichols, 1985).
Well Development
After allowing sufficient time for well seals and grouts to set (usually 24 hours), wells should be developed as soon as possible because:
1. if there is a problem with the well warranting replacement, every effort should be made to identify this need before the drill rig leaves the site; and
2. it is faster and easier to develop a well before silts and clays have begun to harden at the bottom of the screen.
Laboratory Handling
If turbid samples are to be analyzed, the issue of how the laboratory is to prepare the samples must be resolved beforehand. Laboratory procedures should be outlined, with review and approval by a NYSDEC QA/QC officer.

FIGURE 1 - Decision Flow Chart for Filtration of Groundwater Samples Collected for Metals Analysis

References and Additional Reading

Armstrong, R. and Carlton, G.M., 1984. The suction side sample catcher in groundwater quality sampling: Ground Water Monitoring Review, v. 4, no. 4, p. 48-53.

ASTM, Methods for Investigation of Water and Wastewater.

ASTM, 1981, Draft Standard Guide for Sampling Ground Water Monitoring Wells, p. 14.

Barcelona, M.J., et al, 1984, A laboratory evaluation of ground water sampling mechanisms: Ground Water Monitoring Review, v.4, no. 2, p. 32-41.

Claasen, H.C., 1982, Guidelines and Techniques for Obtaining Water Samples that Accurately Represent the Water Chemistry of an Aquifer: U.S. Geological Survey, Open-File Report 82-1024, p. 41-48.

Dragun, J., 1988, The Soil Chemistry of Hazardous Wastes; Hazardous Materials Control Research Institute, Silver Spring, Md., 458 p.

Emenhiser, T.C., and U.P. Singh, 1984, Innovative sampling techniques for ground water monitoring at hazardous waste sites: Ground Water Monitoring Review, v. 4, no. 4, p.35-37.

Gibb, J., Schuller, R., Griffen, R., 1981, Collection of Representative Water Quality Data from Monitoring Wells: Proceedings of the 7th Annual Research Symposium, Land Disposal: Municipal Solid Waste, p. 134-135.

Gibb, J.P., et al, 1981, Procedures for the Collection of Representative Water Quality Data from Monitoring Wells: Illinois Water Survey Cooperative Report No. 7, 61 pp.

Gibb, J.P., Schuller, R.M., Griffin, R.A., Monitoring Well Sampling and Preservation Techniques, EPA-600/ 9-80-101.

Handbook for Sampling and Sample Preservation of Water and Wastewater, Sept. 1976: U.S. Department of Commerce, NTIS PB-259 946.

Ho, J. S-y, 1983, Effect of sampling variables on recovery of volatile organics in water: J. Amer. Water Works Assoc., p. 583-586.

Humenick, M., Turk, L., and Colchin, M., 1980, Methodology for monitoring groundwater at uranium solution mines: Ground Water, v. 18, p. 262-273.

Kennedy, V., Zellweger, G., and Jones, B., 1974, Filter pore size effects of the analysis of Al, Fe, Mn, and Ti in water: Water Resources Research, v. 10, p. 785-790.

Lindsay, W.L., 1979, Chemical Equilibria in Soils, John Wiley and Sons.

Methods for Chemical Analysis of Water and Wastewater, March 1979: EPA-600/4-79-020, USEPA EMSL Laboratory, Cincinnati, Ohio, 45268.

Nacht, S.J., 1983, Monitoring sampling protocol considerations: Ground Water Monitoring Review, v. 3, No. 3, p. 23-29.

New York State Department of Environmental Conservation, New York State Contract Laboratory Protocol; November, 1987, Albany, New York.

Nightingale, H., and Bianchi, W., 1979, Influence of well...and monitoring: Water Resources Bulletin, v. 15, p. 1394-1407.

Pettyjohn, W.A., et al, 1981, Sampling Ground Water for Organic Contaminants: Ground Water, v. 19, no. 2, p. 180-189.

RCRA Comprehensive Groundwater Monitoring Evaluation Document, March, 1988, USEPA-OWPE, Directive 9950.2, p. 22.

RCRA Ground-Water Monitoring Technical Enforcement Guidance Document, 1986, USEPA-OWPE, OSWER-9950.1, p. 114.

Scalf, M.R., et al, 1981, Manual of Ground Water Quality Sampling Procedures: EPA-600/2-81-160.

Schmidt, K.D., 1982, How representative are water samples collected from monitoring wells?: Proceedings of the Second Annual Symposium on Aquifer Restoration and Ground Water Monitoring.

Schuller, R.M., et al, 1981, Recommended sampling procedures for monitoring wells: Ground Water Monitoring Review, v. 1, no. 1, p. 42-46.

Stolzenburg, T.R., and Nichols, D.G., 1985, Preliminary Results on Chemical Changes in Groundwater Samples Due to Sampling Devices, Interim Report; Electric Power Research Institute, EA-4118 Research Project 2485-7.

Strawsberg, S., March/April 1983, Turbidity interferes with accuracy in heavy metals concentrations: Industrial Wastes, p. 20-21.

Test Methods for Evaluating Solid Waste, November 1986, EPA-SW-846, Third Edition, Washington, D.C.

Trela, J.J., 1985, Should water samples from monitoring wells be filtered? Yes and No. From symposia titled: Should Ground Water Samples from Monitoring Wells be Filtered Before Laboratory Analysis? National Water Well Association.

Unwin, J., 1982, A Guide to Groundwater Sampling: NCASI Technical Bulletin No. 362., p. 46-48.

Williams, E.B., 1981, Fundamental concepts of well design: Ground Water, V. 19, no. 5, p. 527-542.