Lecture 8: Designing a water quality monitoring program (incl. QA and QC)

Why monitor water quality?

Without monitoring it is difficult to find the source of water quality issues. Water monitoring information is also used to guide improvements or prevent further degradation.

Monitoring programs

All monitoring programs are different because they reflect the environment and issues being faced.

Rationale for monitoring

• Research

  • test hypothesis (experiment resulted in a significant change)

• Regulatory compliance or enforcement

  • chemical industrial plant does not exceed specified emission

  • effluent characterisation/ reporting

• Response to an event

  • identify specific issues (of significance)

  • identify what and who is responsible (forensic)

• Routine monitoring

  • spatial and temporal trends

• Managing a waterbody

  • can involve any of the above

Knowns and unknowns

• Known knowns (e.g. contaminants that are measured)

• Known unknowns (e.g. contaminant known to be used but are not monitored)

• Unknown unknowns (e.g. contaminants we are not aware of and therefore can not or do not monitor for)

Non-targeted analysis

• Uses advanced analytical instruments such as

  • high performance liquid chromatogram

  • mass spectrometer

  • mass spectrometer (HPLC-MSMS)

  • quadrupole time of flight mass spectrometer (QTOF)

• Any organic chemical can be detected (based on mass of fragments) then you need to identify the chemical from libraries and quantify them

• The objective is to identify chemicals present

• Can also use non-targeted to survey chemicals not already analysed and determine if they should be added to suite of targeted chemicals

Limitation of non-targeted analysis

• Without standard the chemical can only be identified and semi-quantified '

  • you can still determine if the concentration is increasing over time

• Concentration of chemicals can only be determined if a standard is available and a concentration response relationship (calibration curve) has been developed

  • if you know the peak area and have a calibration curve you can determine the conc

Framework for monitoring program design

• Regular review (3 to 5 years) to recommend improvements

  • is the program still fit for the purpose?

  • has the situation changed?

  • could additional stressors or objective be included?

Considerations in a sampling program

• Objectives

  • baseline monitoring?

  • trend detection?

  • search for hotspot?

  • margin of error allowable?

• Variability

  • spatial variation

  • temporal variation

• Cost factors

  • sampling cost

  • analytical costs

  • fixed vs. minimum cost

• Nontechnical factors

  • sampling convenience

  • accessibility

  • availability of resource

  • regulations

Underlying principle of water sampling

Samples must be representative

Developing a conceptual model

• A conceptual model (diagram or figures) of the area to be monitored includes:

  • known stressors

  • transport pathways

  • organisms that may be affected and how

• They express ideas about components and processes deemed important in a system, document assumption about how they are related and identify knowledge gaps

Examples of conceptual models

• Conceptual models

• Qualitative models

Site selection

• A site monitors water passing through that point

  • water is a combination of water from all land and land uses upstream of site

• Thus, site selection must reflect:

  • questions to be answered, objectives and hypothesis to be tested

  • processes that reflect water quality

  • natural and man-made features

  • an area where water is well-mixed and representative (i.e. not stratified, still or slow moving)

  • good logistics (i.e. accessibility, susceptibility to flooding, availability of electricity)

Environmental Protection (Water and wetland biodiversity) Policy 2019

• The purpose of this policy is to achieve the objective of the Environmental Protection Act in relation to waters and wetlands by:

  • identifying EVs for waters and wetlands

  • identifying management goals for waters

  • stating WQGs and objectives to enhance or protect the environmental values

  • providing a framework for making consistent, equitable and informed decisions about waters

  • monitoring and reporting on the condition of waters

Quality Assurance and Quality Control (QAQC)

QAQC is essential to the success of a monitoring program. If the data is not accurate then wrong conclusions, decisions and investments may be made.

Pros and cons of grab samples

• Pros

  • can collect many samples

  • can monitor temporal and spatial changes of contaminant concentrations

  • cheaper per sample

  • manual grabs have high labour costs but low capital and maintenance costs

  • automated grabs have high capital and maintenance costs but low labour costs

• Cons

  • measure the conc at a point in time (could miss contaminants)

Pros and cons of passive samplers

• Pros

  • measure average aqueous conc during deployment

  • conc chemicals from water — can detect more chemicals at lower aqueous concentrations

• Cons

  • cannot monitor short-term temporal variation of contaminant concentrations

  • more expensive per sample

  • a lost sampler means no data for the deployment period (e.g. one month)

Real-time monitoring with probes

• Measure pH, dissolved oxygen, temperature, conductivity, turbidity, nitrogen and phosphorus, discharge (direction and velocity of flow), pesticides (not yet), toxicity (possible but complicated)

• The sampling frequency can be selected from secs to weeks (this reduces sampling labour)

• Need to develop relationships between real-time and
  traditional measurements, potential role of real-time
  measurement to provide estimates of traditional measures

Spatial considerations

• Avoid sampling where the following occur:

  • riffles

  • backwaters

  • confluences (200 m downstream on major rivers)

• Stratification occurs in little a 2 m

• Tidal mixing zones (fresh over salt)

• High-low temperature

• Stagnant waters (lakes and dams)

• Plume buoyancy (density/temperature issue)

• Always collect samples from control sites

Temporal considerations

• Tidal fluctuations

  • always collect sample on outgoing (ebb) tide

  • river mouth to upstream, otherwise you might be following the water downstream (i.e. collecting the same water)

• Releases

  • e.g. on an ebb tide as for some STP discharges

• historical monitoring points

  • EHP/ other agencies

  • Papers/reports/universities

  • Facility under investigation

• Diurnal changes

  • e.g. DO

Identifying background sources of contamination

• E. coli and enterococci are used to indicate contamination from human faeces but could be the result of birds, wildlife and livestock

  • differentiate using DNA test

• Nutrients (ammonia, nitrogen and phosphates) can be introduced from farms, gardens, constructure activities

• Be aware of stratified water bodies

  • bottom layers may have anoxic conditions which release bioavailable nutrients from sediments

• Total Petroleum Hydrocarbons (TPH) and zinc introduced from roads

  • this is why it is important to take control samples

Quality Assurance (QA)

Preventing quality issues through written processes and implementation of these processes, training, audits and what tools are used. A proactive approach includes root-cause analysis to investigate errors so that we can modify processes to ensure it is not repeated (e.g. staff training programs, calibration processes, written procedures and record keeping).

This also includes a quality control program.

Quality Control (QC)

The focus is on testability, checking and testing results (e.g. making sure there is no contamination in field gear, that laboratories are producing good quality results, that sampling processes are fit for purpose, sampling coverage is adequate to address quality needs).

Precision vs. accuracy

• Precision and accuracy are two ways to think about error

  • accuracy — how close a measurement is to an accepted value/tolerance

  • precision — how close measurements of the same item are to each other

• Note that precision is independent of accuracy

Sampling plan

• Select the correct analytes, understand holding times, preservation etc.

  • consider correct type and number of sampling containers and analysis

  • adhere to correct holding times and preservation

  • consider lab quality control requirements and accredited labs/tests

  • prepare field filtration required (ammonia, dissolved metals, DOC)

  • do samples need to be filled to ‘zero headspace’?

  • consider storage and transport methods

• Consider Limits of Reporting (LOR) or Practical Quantification Limits (PQL)

• Obtain a representative sample/s of a water body

• Account for variability

  • temporal and spatial variation

  • understand background influences

• Collection evidence

Limits of Reporting (LOR) or Practical Quantification Limits (PQL)

Limits of Reporting (LOR) are the lowest concentration that the laboratory can quantify a chemical with certainty. The LOR of the laboratory should be below the guideline value for the analytes of interest.

Replicates to assess variability

• Water bodies are not homogenous temporarily or spatially but typically heterogeneous

• Three replicates across a stream at different depths (minimum)

  • good approach if sampling requires high defensibility or variability needs to be assessed

• In situ readings can indicate if a waterbody is well mixed vertically

Sample ID

• Unique ID should include:

  • date

  • location description

  • sample number

• Always ensure that:

  • samples names are not too long

  • naming system is consistent and suitable

  • use waterproof marker and write label before taking sample (avoid sharpie PFAS)

Types of QC blanks

• Container blank

  • used to quantify and trace contamination problems associated with the sample containers and preservation

• Trip or transport blanks

  • contamination due to shipping and laboratory sources;

  • usually for Volatile Organic Carbons (VOCs) analysis and when using ice.

• Equipment blanks

  • contamination due to intermediate sampling tools (bucket, bailer etc.)

• Field blanks

  • contamination due to field conditions, filtration, transport and container

*only ultra-pure water provided by the analysing laboratory should be used for blanks

Multiparameter Water Quality Meter

• Always be calibrated pre and post sampling (i.e. dissolved oxygen, electrical conductivity, pH) particularly for investigations:

  • calibrate the meter for expected EC range (freshwater/brackish/marine)

  • EC varies with temp; hence data are corrected to 25°C (specific conductance)

  • check instrument to see if it corrects EC

• Keep moving probe if membrane – needs flow), luminescent

• Meters should be occasionally cross-checked using another meter

• Record calibration details (defence may request records) in a meter logbook with:

  • maintenance

  • repair

  • calibration

• Ensure suitable supplies of standard calibration solutions are maintained and shelf-life is monitored

Filtering

• Ensure syringes/filters are new and appropriate for analysis

  • cellulose acetate filters are used for metals

  • polyethersulfonone (PES) for nutrients

  • use 0.45 micron for dissolved ammonia/metals

  • do not use glass fibre pre-filters for metals

  • use filters that have been sealed

  • check expiry dates

  • perform blank checks

Equipment preparation

• Prepare sample bottles (group for ease of use)

• Have documents, record keeping logbooks ready:

  • appropriate Chain-of-Custody documentation

  • appropriate check lists and contemporaneous notes

• Have ice-bricks and/or ice ready, dry ice if freezing samples

  • eskies on standby

• YSI meters correctly maintained, calibrated and working

  • calibration solutions should be present if away overnight

Sample bottle information

Container, storage and transport

Observations

• Important to capture as much detail on both incident location and waterway as possible

• Officers should take GPS waypoints of each sample site

• Photos

• Videos

• Records of general observation (contemporaneous notes):

  • water colour and clarity

  • odours

  • water flow, velocity/direction

  • weather conditions

  • surface scum

  • algal/plant growths

  • dead or dying fish

  • proximity to earthworks, construction, agricultural, industrial or water treatment activities, weirs '

  • maps or drawings

In situ measurements

• In situ measurements should be taken at all sample locations first

  • gives an indication of stratification

• Turbidity should be measured in situ

  • probe should be 0.3 m below surface, but use middle of water column for very shallow waters, the 1 m intervals in fresh, 2 m in estuaries

  • readings should be taken as close to middle of channel as possible

  • probe should be kept in gentle motion whilst the meter stabilises

  • record mg/L and % for dissolved oxygen

  • pH (holding time 6 hours) and turbidity/conductivity (holding times 24 hours) can be done in lab but in situ preferred

Chain of Custody (COC)

A legal term used to describe the ability to guarantee the identity and
integrity of the sample from field sampling through to receipt at the lab. A process used to preserve and document the chronological history of a sample from person-to-person because test results are regularly used as evidence in legal proceedings (disprove tampering or unintentional misconduct). Breaks in COC discredit data.

Checking data

• Data should be checked ASAP after receiving results

  • check original field sheets vs lab report (time, date, location, site IDs)

• Sign off QC’ed data once passed

• Check blanks for contamination

  • field blank (FBLK) concentration values should be equal to or less than two times the corresponding Practical Quantitation Limit (PQL) values

  • trip blank (TBLK) concentration values should be equal to or less than two times the corresponding PQL values

Relative Percent Difference (RPD) for duplicates

RPD of ≤ 20% is an acceptable result for duplicate aqueous samples, provided the result is 5-10 x the LOR. Where the result is close to the LOR, RPD may exceed 20%.

The Australian Standard for Soils (AS 4482.1-2005), which can be applied to sediments, gives an acceptance criteria RPD of 30-50%