R. Pitt January 4, 2007

Module 3: Water Sample Collection Methods

General Considerations for Sample Collection ............................................................................................. 1 Basic Safety Considerations when Sampling ............................................................................................................1 Selecting the Sampling Locations .............................................................................................................................2 Volumes to be Collected, Container Types, Preservatives to be used, and Shipping of Samples .............................3 Sample Volumes....................................................................................................................................................5 Sample Containers.................................................................................................................................................6 Cleaning Sample Bottles........................................................................................................................................7 Field Processing of Samples and Preparation for Shipping ...................................................................................8 Shipping Samples ..................................................................................................................................................9 Chain-of-Custody and other Documentation .........................................................................................................9 Sample Preservation and Storage at the Laboratory ............................................................................................10 Personnel Requirements ..........................................................................................................................................10 Receiving Water and Discharge Sampling ................................................................................................. 11 Automatic Water Sampling Equipment ...................................................................................................................11 Required Sample Line Velocities to Minimize Particle Sampling Errors............................................................18 Programming an Automatic Sampler...................................................................................................................18 Automatic Sampler Initiation and the use of Telemetry to Signal or Query Sampler Conditions .......................23 Manual Sampling Procedures ..................................................................................................................................24 Dipper Samplers ..................................................................................................................................................24 Depth-Integrated Samplers for Suspended Sediment ..........................................................................................25 Bed-Load Samplers .............................................................................................................................................26 Floatable Litter Sampling ....................................................................................................................................27 References .................................................................................................................................................. 28

General Considerations for Sample Collection This module summarizes some of the most important issues associated with sample collection. A more detailed discussion is presented by Burton and Pitt (2002).

Basic Safety Considerations when Sampling The most important factor when conducting a field monitoring program is personnel safety. If an adequate program cannot be carried out in a reasonably safe manner, then an alternative to the monitoring program should be used. Most of the hazards reflect site selection and sampling times. The use of automatic samplers and well trained crews (more than one person in the field!) will reduce many of the hazards. Sampling may expose field personnel to hazardous conditions. Obviously, water hazards (high flows, deep pools, soft sediments, etc.) are usually of initial concern. In some studies, sampling during rainy weather in streams that may undergo rapid velocity and depth changes is necessary. Great care must be taken when approaching a stream in wet weather, as steep and slippery banks may cause sliding into the water. Always sample in pairs and have adequate safety equipment available. At a minimum, this will include: • throw rope • inflatable life vests

• nylon covered neoprene waders (that offer some floatation, even when swamped) • 2-way radio or cellular phone • weather radio If the conditions warrant (such as with steep and slippery stream banks), the sampler personnel should be tied together, with an attachment to a rigid shore object. In all cases, only go into the stream if absolutely necessary. Try to collect all samples from shore, especially if during heavy rains. Be extremely cautious of changing weather and stream conditions and cancel sampling when hazardous conditions warrant. Never enter a stream where footing is unstable or if the water is too deep (probably more than 2 feet deep) or fast (probably more than 2.5 ft/sec). Always enter the water cautiously and be prepared to make an efficient retreat if insecure. Other hazardous conditions may also occur when working near urban streams. Sharp debris in the water and along streams require protective waders be worn at all times while in the stream. No one should enter the water barefooted. Poison ivy, poison oak, and ticks thrive along many stream banks, requiring long pants and shirts. When in the field during sunny weather, sun screen and a hat are also necessities. In many parts of the country, especially in the south, special caution is also required concerning snakes. Water moccasins are very common and coral snakes and copperheads may also be present along streams. Again, waders offer some protection, but be careful when moving through thick underbrush where visibility is limited. These cautions are necessary and are basically common sense. However, the greatest dangers associated with field sampling, especially in urban areas, are likely associated with loose running dogs, weird people, automobiles/trucks, and eating greasy fast food (dangers which are not restricted to stream sampling).

Selecting the Sampling Locations Specific sampling locations are determined based on the objectives of the study and site specific conditions. Obviously, safety is a prime consideration, along with statistical requirements expressed in the experimental design. In all cases, the sample must represent the conditions being characterized. The process of selecting a sampling site is often given minimal thought when designing an assessment study. Site selections are driven by two basic criteria: accessibility/safety and upstream-downstream locations of pollutant discharges. Channel, flow, and stratification characteristics are particularly important when locating sample sites in streams, rivers, lakes and reservoirs. Sampling near shore is seldom satisfactory except in small, upper reach streams. Whether using a random or systematic approach, one should carefully note the channel, flow, or stratification (lakes and reservoirs) conditions. In reservoirs it is common for the principal flow to follow the old river channel and at a depth similar to the temperature (density) of the feeder stream. This area thus often contains the highest pollutant concentrations (e.g., suspended solids, fecal pathogens). Depositional zones, such as river bends and mouths, pools, and impoundment structures, should be sampled for sediment contamination and toxicity. As noted previously, paired analyses are the most efficient sampling strategy. This can be simply sampling the influent and effluent of a control structure, outfalls of test and control watersheds, comparable stream habitats in test and control streams, or even the same stream sampling location, but at different seasons. Paired sampling can eliminate much variability, as many influencing factors are assumed to remain constant, enabling effects to be more easily seen. Obviously, if the expected differences are expected to be large between the two elements in the pair, and the background random variability is small, many fewer sampling pairs are needed to identify a statistically significant difference in the observations. Great care must be taken to select correct pairs, as the random variability can easily be greater than expected. One example of likely inefficient paired sampling is sampling above and below an outfall in a stream. In almost all cases, the stream pollutant loads and flows are much greater than a single outfall discharge. Therefore, the differences expected in stream water quality upstream versus downstream of an outfall would be very small and very difficult to detect. Exceptions may occur with large point source outfalls discharging during very low flow conditions. Otherwise, one large number is basically subtracted from another large number (with both having a certain amount of uncertainty) to determine the effects of a relatively small discharge. If this sampling strategy needs to be employed, make sure that the outfall discharge is also well characterized.

The actual location of sampling is somewhat dependent on the type of sampler to be used. However, in all cases, the sample taken must be representative of the flow to be characterized. Permanently mounted automatic or semiautomatic samplers are most restricted in their placement, as security and better access is needed than with manual grab sampling. With manual sampling, less equipment is generally being carried to the sampling location (some type of manual dipper sampler, plus sample bottles, for example), while automatic samplers require a relatively large sample container, a multi-bottle sampler base, and batteries and other maintenance and cleaning supplies to be periodically carried to the sampler. Weekly visits to automatic samplers, at least, are typically needed for maintenance. In all cases, access during rains must be provided to all stormwater sampler locations. Manual stormwater sampling takes place during the rains, of course, while automatic samplers may need to have their bottles switched during rains, or other checks made. Therefore, dangerous locations, such as requiring steep ascents down clayey stream banks must obviously be avoided, for example. Obviously, care must also be used to locate the sampler intakes to minimize induced scour of sediments and to prevent clogging from debris. All submerged pumps can quickly fail if the pump draws coarse particles into the pump, but doesn’t have enough velocity in the sample line to discharge most of them completely through the sample line. If the intake is located on a creek bottom, the water entering the sampler intake will likely scour sediment from the surrounding area. Locating the sampler intake on top of a small anchored concrete slab in the creek will minimize scour. Elevating the sampler intake above the creek bottom will also minimize scour, but would present an obstruction to flows and would easily catch debris. Elevating the intake slightly would be important to obtain a better sample if the flow is vertically stratified. In some cases, sampler intakes can be successfully located on the downstream side of a bridge piling or pier. Do not locate the sample intake near any treated wood structure if heavy metals or organics are to be sampled. Locating a sampler intake in an outfall pipe presents other problems. Because the pipe is likely to be smaller than a receiving water, horizontal differences in water quality should not be a problem. However, vertical differences may occur. The sampler intake also presents a greater obstruction to the pipe flow and therefore has a greater tendency to catch debris. To ensure a well-mixed water sample, the intake should be placed in an area that has turbulent flow. This may decrease volatile components in the water sample, but typical automatic samplers are inappropriate for collecting samples for volatile analyses anyway. Locating the intake on the downstream side of a flow monitoring flume would help obtain a mixed sample. In addition, added obstructions (bricks and concrete blocks) can be cemented to the pipe above the sampling location to induce well-mixed conditions during low to moderate flows. Obviously, flow measurements would not be taken where obstructions are used to mix the flow. Manual sampling is much more flexible and can be modified to better represent the flow conditions at the time of sampling. Obviously, multiple dips across a stream, and at multiple depths will result in a better representation of the stream than a single sampling location. Special manual samplers are needed to collect depth-integrated samples that may be needed for sediment transport studies. The advantages of manual sampling compared to automatic sampling are off-set by the time frame that is represented in the sample. A grab sample taken at a single time will not be as representative of a storm event as an automatic sampler taking sub-samples from many time periods during the event, even considering multiple versus single sampling points. A single sampling location will be subjected to varying conditions during the storm, including variable horizontal and vertical variations. However, if a single sampling location is consistently biased compared to the cross-section of the stream, then that needs to be recognized and corrected. Therefore, it is necessary to observe conditions in the stream during the sampling times as much as possible to detect any potential bias. A bias may be caused by currents or nearby discharges, for example, and may be visually observed if colored or turbid water is indicating current conditions near the sampler.

Volumes to be Collected, Container Types, Preservatives to be used, and Shipping of Samples The specific sample volume, bottle type, and preservative requirements will be specified by the analytical laboratory used. Standard Methods for the Examination of Water and Wastewater lists the basic container requirements, minimum sample sizes, required preservative, and the maximum storage period before the analyses need to be conducted. Table 1 shows these guidelines for water samples. Care must be taken to handle the samples properly to

ensure the best analytical results. Numerous losses, transformations, and increases in pollutant concentrations may occur if these guidelines are not followed. Some analyses should be conducted as soon as possible (within a few hours of sample collection, or preferably on-site or in situ). These include: CO2, chlorine residual, DO unless fixed, iodine, nitrite, ozone, pH, and temperature. ORP (oxidation reduction potential) is also in this category of required on-site analyses, even though not on this table. Parameters that need to be analyzed within 24-hours of sample collection (same day) include: acidity, alkalinity, BOD, cyanide, chromium VI (and other specific ionic forms of metals), taste and odor, and turbidity. Most of the nutrients need to be analyzed within 2 days. Many parameters can be stored for long periods of time, after preservation, specifically total forms of most heavy metals (6 months) and extracted organic compounds (30 days). In some cases, it may be possible to deviate from these guidelines if sitespecific testing is conducted to demonstrate acceptable pollutant stability. The most important guidelines are the bottle type and preservative. Some parameters may be able to undergo longer storage periods, but this must be tested for specific conditions. The required sample volumes are all much greater than needed for most modern laboratory procedures and may be reduced if shipping costs or sample storage facilities are of a concern. Make sure that extra sample is available to redo critical analyses if problems develop, however.

Table 1. Summary of Special Sampling and Handling Requirements for Water and Wastewater Samples th 1 (Standard Methods 19 Edition, 1995; Collection and Preservation of Samples, 1060) 2

4

Sample 3 Type

Preservation

P, G(B) P, G P, G P P, G G

Minimum Sample Size mL 100 200 1000 100 100 100

G G G G, C G, C G, C

Carbon dioxide COD

P, G P, G

100 100

G G, C

Chloride Chlorine, residual Chlorine, dioxide Chlorophyll Color Conductivity Cyanide: Total

P, G P, G P, G P, G P, G P, G P, G

50 500 500 500 500 500 500

G, C G G G, C G, C G, C G, C

Fluoride Hardness Iodine Metals, general

P P, G P, G P(A), G(A)

300 100 500 500

G, C G, C G, C G

P(A), G(A) P(A), G(A)

300 500

G G, C

Refrigerate Refrigerate Refrigerate None required None required Analyze immediately; or refrigerate and add H3PO4 OR H2SO4 TO pH