Department of Health Seal

TGM for the Implementation of the Hawai'i State Contingency Plan
Section 10.6


More than one type of field QA/QC sample may be collected simultaneously to provide a broad assessment of sample data quality. Field QA/QC sampling is typically used to evaluate the following, as well as other considerations:

  • Accuracy of sample collection, processing and analysis procedures through the use of field replicates.
  • Accuracy of sample analysis procedure through the use of field replicates sent to separate laboratories.
  • Effectiveness of sample collection equipment decontamination procedures through the use of equipment blanks.
  • Sample handling and transportation procedures through the use of trip blanks– for samples in aqueous media being analyzed for volatile chemicals.

The frequency of QA/QC sample collection is strongly dependent upon a variety of factors including the sample matrix (i.e., soil, water, or gas phase), COPCs, and QA/QC questions to be answered. The number of field QA/QC samples is site and project specific and needs to be addressed in detail in the SAP or QAPP for each project. A QA/QC sample should be analyzed for the same constituents and by the same method as the primary sample.

QA/QC samples should be labeled in a manner that does not allow the analytical laboratory to identify or correlate the QA/QC sample to the primary sample. This is often referred to as the submission of "blind" samples to the analytical laboratory. For example, if only one primary sample is collected, a replicate QA/QC sample label should not identify it as a duplicate (or other) QA/QC sample. If more than one primary sample is collected, the QA/QC sample name may indicate that it is a duplicate (or other) QA/QC sample as long as correlation to the primary sample is not possible (e.g., by indicating a sample collection date or time different from the primary sample).

10.6.1 Field Replicates

Field replicate samples are duplicate or triplicate samples collected from within the same decision unit or from the same groundwater well to evaluate the precision of the sampling effort. Replicates are to be collected, preserved, stored, transported, and analyzed in the same way as primary field samples. Duplicate or triplicate samples are collectively referred to as "replicate" samples unless specifically indicated. Replicates are intended to represent the same population and are taken to provide information on precision, accuracy, and representativeness for the data collection activity (e.g., replicates provide a measure of contaminant heterogeneity for a specific decision unit). If the degree of contaminant heterogeneity exceeds established DQO in the SAP, then additional sampling and/or steps to limit errors during sample processing and analysis are typically required to provide representative sample data. A field replicate precision of approximately 10-35% is generally established as a DQO, depending on the media and contaminant.

The method for collecting replicate QA/QC samples is strongly dependent upon the sample matrix, the COPCs, and the QA/QC questions to be answered, particularly when considering Multi-Increment sample (MIS) techniques as compared to discrete sampling. Discrete Sampling Replicates

In general, for discrete sampling of groundwater or soil the HEER Office recommends collecting one replicate QA/QC sample per field day per sample matrix, or 10% of all field samples, whichever is greater. At least 10% replicate QA/QC samples should be collected in each decision unit or each area of known or suspected contamination. Consider both the horizontal and vertical dimensions when planning replicate QA/QC sample locations. If small scale heterogeneity is expected at the site, additional replicate QA/QC samples may be required to assess the scale of heterogeneity. Different project-specific frequencies may be proposed to best meet project DQO. If proposing different QC sampling frequencies for a specific investigation, the proposed QC sampling program and the rationale should be presented in detail in the project-specific SAP or QAPP and discussed with the HEER Office prior to field investigation.

Co-located duplicate samples

Co-located duplicates are samples collected at the same time from a location in proximity to the primary sample. Co-located duplicate soil samples are commonly collected due to sample volume factors (i.e., the volume of sample material retrieved in the sampler is less than the volume of sample required for laboratory analysis). Minimize the distance between the primary and duplicate sample collection points; small scale heterogeneity in the contaminant distribution is more likely as the distance increases.

The co-located samples would be expected to have similar contaminant concentrations. Data quality objectives to evaluate the precision of co-located samples should be included in the SAP or QAPP, and co-located sample data compared to ensure these DQO are met.

Duplicate groundwater samples

For non-volatile groundwater contaminants collected in vials, generally two sample containers are "alternately" filled. For example, if a low-flow pump is used, the two containers would be filled by going back and forth with the discharge tubing.

For volatile groundwater contaminants, where multiple 40 ml vials are commonly used for each sample and loss of volatiles is an important concern, the primary and duplicate sets of samples are collected alternately. One vial is completely filled with the primary sample then a duplicate sample vial is filled until all vials (primary and duplicate sets) for that one sample are collected. It is also important to follow vial filling protocols appropriate to ensure minimal agitation and zero-headspace for the volatile samples.

Relative Percent Difference

In certain cases, particularly for discrete sampling, only duplicates rather than triplicates may be available to evaluate precision of sampling data, though triplicates are recommended wherever feasible. In those situations where only duplicates are able to be collected, the precision of the data would be evaluated by determining the RPD.

The RPD is calculated as described in Subsection 10.3.1. Multi-Increment Soil Sampling Replicates

The Multi-Increment soil sampling approach relies on collection of field replicate (triplicate) samples to estimate the sampling precision, as discussed in Section 4.2.3. Collecting and analyzing triplicate samples allows for statistical calculation of several important quantities including the standard deviation, RSD, and 95 percent (%) upper confidence level (UCL) of the mean. These statistical evaluations are utilized to determine the degree that the measured levels of contaminants vary from the (estimated) mean, and is taken into consideration when comparing site data to applicable HEER Office EALs (See Section 4.2.5).

The number of decision units where Multi-Increment sample replicates are collected will vary with each project, total number of decision units, and site characteristics. Consequently the number of DUs with replicates is site-specific and determined as part of the overall sampling strategy in the SAP. A batch-type replicate approach (similar to that used in the lab) may be applied in the field, if multiple decision units are similar (e.g., similar soil type, contaminants of concern, history of chemical use, topography, etc.). If multiple similar DUs are evaluated on a site, replicates in one DU may be used to evaluate that DU and up to 9 similar DUs. In this case, the precision data determined for contaminant(s) in one DU (e.g., RSD) would also be applied to the other DUs in the similar batch.

Standard Deviation, Relative Standard Deviation, and 95% Upper Confidence Limit of the Mean

The standard deviation is a statistical measure of the scatter, or variability, of several sample values around their mean (or average). The lower the standard deviation, the lower the variability of the sample values observed in the data. The standard deviation may be informally interpreted as the size of a "typical" deviation from the mean (or average) and may be calculated using standard equations presented in an introductory statistics book or included as software functions in programs such as Microsoft Excel.

The RSD, expressed as a percent, is a measure of precision among several sample values (the normal, duplicate, and triplicate samples in the case of Multi-Increment sampling). The RSD differs from the RPD in that it measures the precision among several sample values versus between just two sample values. The RSD can be calculated as the standard deviation of the sample replicates divided by the mean (or average) of the sample replicates, times 100%.

An RSD of 35% or less is typically a goal during environmental investigations. However, an RSD greater than 35% does not necessarily mean the data is not usable for the intended purpose. For example, an RSD somewhat greater than 35% may be acceptable if the estimated average level of contaminant(s) in the DU is much greater or much less than the relevant HDOH Tier 1 EAL.

The 95% UCL is another statistical measure of the precision for a series of sampling measurements. In this case, the normal, duplicate and triplicate samples are used to calculate a mean (or average) value and a standard deviation. The mean and standard deviation are used to calculate, with 95% confidence, the mean value for the individual decision unit. Formulas and spreadsheets for calculating the 95% UCL are available through websites providing statistical analysis support.

10.6.2 Blanks

Blank QA/QC samples are aliquots of the sample matrix that is known to be free of contaminants. The analytical data for blanks provides a measure of the cross-contamination that may have occurred during sample collection, sample storage and transport, or during laboratory preparation, extraction, and analysis. Compare the analytical results of the various types of blanks to each other to assess the degree to which contamination may have been introduced into the samples. Trip Blanks

The purpose of a trip blank is to assess the possibility of cross contamination during sample collection, storage, and transport to the analytical laboratory. The trip blank is typically analyzed for volatile organic compounds in aqueous samples due to the high vapor pressure and potential for vapor migration. Non-aqueous samples collected using methanol preservation techniques may also require a trip blank.

Prepare trip blanks by filling sample containers with reagent grade water, then assuring that the trip blank sample containers accompany the main sample containers along every step to the analytical laboratory. Trip blanks are not opened in the field. Trip blank water should be from the same source as the method blank water used in the laboratory. Field Equipment Rinsate Blanks

The purpose of an equipment blank (also commonly referred to as a field equipment rinsate sample) is to place a mechanism of control on sample collection equipment (i.e., soil core samplers or sample tubing) that is decontaminated and reused in the field. Specifically, an equipment blank assesses sample collection equipment and/or related ambient conditions that may affect sample quality. Because the equipment blank is stored and transported with the primary samples, it is also representative of sample bottle preparation, storage, and transport conditions.

An equipment blank is collected by pouring reagent grade water over/through decontaminated equipment used in sample collection. The water is then collected in a sample container and analyzed for the contaminants of interest. Equipment blank water should be from the same source as the method blank water used in the laboratory.

The use of field equipment rinsate blanks is important for ultraclean and very low level (trace) contaminant investigations; however, in many general contaminant investigations it is not necessary as long as a specific and effective protocol (i.e., SOPs) for field decontamination of any re-used sampling tools is documented in the SAP and utilized. Collection of large Multi-Increment soil samples (rather than discrete samples) further decreases the potential for cross contamination with trace amounts of soil left on a sampling tool.

The protocol for decontamination should ensure that new sampling equipment is decontaminated (or certified clean and in original container until used) and any previously used equipment is decontaminated before reuse. The SAP or QAPP should clearly identify if the site investigation will or will not include equipment rinsate blanks, and discuss the rationale for this decision. Field (water) source blanks are required to be analyzed whenever equipment is decontaminated in the field. Where equipment rinsate blanks are included for trace level investigations or for other reasons, the HEER Office recommends collecting one equipment blank per matrix per sampling team per day. Field Source Blank

Field source blanks are collected from the water source used for decontamination rinse of equipment, and are used to assess potential for contamination in the water used for decontamination. One source blank is collected from each source of water used for decontamination.

10.6.3 Documentation

Document the following sampling information, as applicable, for primary and QA/QC samples in the field log:

  • Time and date of sample collection
  • Name of person(s) collecting the sample
  • Location of sample
  • Sampling procedure
  • Sample identification
  • Source of blank matrix
  • Table that provides a cross reference of primary and replicate samples
  • Equipment decontamination procedure

10.6.4 Chain of Custody

Attach a label to the sample jars and log each sample on a chain-of-custody (COC) form. Provide at a minimum the following information on the COC:

  • Project identification
  • Samplers name
  • Sender - company name and address
  • Destination - laboratory name and address
  • Sample identification
  • Number of sample containers per sample
  • Preservation, if any
  • Date and time of sample collection for each sample
  • Requested analytes
  • Special handling requirements, if any
  • Shipping company
  • Name and signatures of persons relinquishing custody
  • Date and time when custody is relinquished
  • Signatures of persons receiving custody
  • Date and time when custody is received

The chain-of-custody must not be broken between the sampler and the laboratory sample receiving personnel. Enter the name of the shipping company into the received custody section, if the samples need to be shipped.