Question 3 of 5: Field-to-Laboratory Sampling Protocol

Question 3: Field-to-Laboratory Sampling Protocol (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

An accurate, defensible environmental sampling result depends on a documented protocol that spans every step between the field and the laboratory report, because an error introduced at any single step — collection, preservation, handling, or data entry — propagates uncorrectably into the final reported concentration. No amount of laboratory precision can recover a sample that was collected, preserved or documented incorrectly.

The sampling protocol. The protocol is settled on paper before anyone goes to the field, and it begins with the objective — a systematic planning step (the Data Quality Objectives process) that fixes what decision the data must support, what contaminants and media are of concern, what concentration threshold matters, and what error rates are tolerable. Those decisions determine everything downstream and are recorded in a written Sampling and Analysis Plan (SAP), comprising a field sampling plan and a quality assurance project plan. The SAP specifies the sampling design and its statistical basis (simple random, stratified, systematic grid, judgmental or composite — see Question 1a); the number of samples, which is fixed by the variability of the medium and the precision the decision requires, not by convenience or budget alone; the exact locations and depths, with the coordinate system and survey method used to fix them; the timing and frequency, including whether the medium is subject to seasonal, diurnal or flow-driven variation; the analytes and the analytical methods, with their required detection limits confirmed to be below the applicable regulatory criteria; the container, preservative and holding time for each analyte; the QC samples and their frequency; the decontamination procedure; and the documentation and chain-of-custody requirements. It also names the responsible personnel, their training and certification, and the site health and safety plan. Sampling to a written, pre-approved protocol — rather than deciding in the field — is what makes the resulting data both statistically defensible and legally admissible, because it prevents the sampling decisions from being influenced by what the sampler sees or expects to find.

Sample collection methods. The collection method must match the medium and the question being asked. A grab (discrete) sample captures conditions at one point in space and time and is appropriate when temporal or spatial variability is itself the object of study (e.g. a spill investigation). A composite sample physically combines several grab samples (time-composited over a shift, or space-composited across a site) to give an economical average concentration, at the cost of losing information about the variability among the individual grabs. For groundwater, a low-flow purge-and-sample technique minimizes drawdown-induced turbidity and cross-contamination between aquifer zones. For soil, sampling must specify a consistent depth interval and a defensible spatial pattern (systematic grid, judgmental/biased toward visible impact, or stratified by land use); for air, the choice between an integrated sampler (drawing air through a sorbent or filter over hours) and a grab/canister sample depends on whether a time-weighted average or an instantaneous concentration is required. In every medium, equipment must be decontaminated between locations (or dedicated/disposable equipment used) to prevent cross-contamination carrying a high concentration from one sample into the next.

Sample preparation and preservation. Once collected, every sample must be placed in a container and preservative appropriate to the analyte (e.g. amber glass and 4°C for organics to prevent photodegradation and volatilization; acidification to pH<2 with HNO₃ for dissolved metals to prevent adsorption/precipitation on container walls; sodium thiosulfate to quench residual chlorine before a bacteriological hold). A holding time — the maximum interval between collection and analysis within which the result is considered valid — is specified for every analyte/preservative combination and must never be exceeded. Samples are kept on ice in coolers during transport (typically ≤4°C for most parameters, or frozen for some), and the specific container material (glass vs. polyethylene) is chosen to avoid sorption or leaching interactions with the target analyte.

Quality assurance and quality control (QA/QC). QA is the overarching program — written Standard Operating Procedures, a documented sampling and analysis plan (SAP), trained/certified personnel, and a certified (accredited) laboratory — that gives confidence the data will be of known, adequate quality. QC is the set of specific technical checks executed alongside every sampling event: field blanks (clean water/media taken through the same handling steps to detect contamination introduced by sampling equipment or the ambient environment), trip blanks (accompany volatile-organic samples through transport and storage without being opened, to detect contamination during shipping), equipment/rinsate blanks (rinse water from decontaminated equipment, to verify decontamination effectiveness), field duplicates (a second, independent sample from the same location, to quantify combined field-plus-lab precision), and matrix spikes (a known quantity of analyte added to a sample split, analyzed to verify the lab's percent recovery is within an acceptable range). A strict, unbroken chain of custody record accompanies every sample from collection to final disposal, documenting every person who handled it.

Data management. Field observations (time, location coordinates, weather, field-measured parameters such as pH/temperature/conductivity, any deviations from the SAP) must be recorded contemporaneously in a bound field logbook or electronic field data system, not reconstructed afterward. Laboratory results are reported with their associated QC data (blank results, duplicate RPDs, spike recoveries, method detection limits) so a data user can independently judge data usability, not just accept the reported number at face value. Results below the detection limit must be flagged as censored (non-detect) data rather than silently treated as zero, and the full data package should be reviewed (data validation) against the QA/QC acceptance criteria before it is used for any decision.

Sources of error. Errors can be grouped into three families. Sampling error arises because a finite sample can never perfectly represent a heterogeneous population — this is reduced (not eliminated) by an adequate, statistically defensible sampling design and sample size. Systematic (bias) error comes from a consistent, correctable source — contaminated equipment, an uncalibrated field instrument, sample degradation from an exceeded holding time, or a non-representative (biased) site selection — and is detected through the blanks, duplicates and calibration checks described above. Random (analytical) error is the inherent variability of the measurement process itself, quantified through replicate analyses and reported as the method's precision. A rigorous protocol addresses all three: a defensible design and adequate sample size control sampling error, disciplined field procedures and QC samples control and detect bias, and laboratory replicate analysis and statistical process control quantify random error — together giving a result that can be defended as an accurate representation of the actual contaminant levels present.